Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Optimal markers for the identification of Colletotrichum species

Optimal markers for the identification of Colletotrichum species bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1 Optimal markers for the identification of Colletotrichum species 3 Willie Anderson dos Santos Vieira 4 Priscila Alves Bezerra 5 Anthony Carlos da Silva 6 Josiene Silva Veloso 7 Marcos Paz Saraiva Câmara 8 Vinson Patrick Doyle 10 Universidade Federal Rural de Pernambuco – UFRPE. Departamento de 11 Agronomia. Rua Manuel de Medeiros, s/n - Dois Irmãos, Recife – Pernambuco, 12 52171-900. 13 Department of Plant Pathology and Crop Physiology, Louisiana State University – 14 LSU, AgCenter, Baton Rouge, Louisiana, United States of America, 70808. 16 E-mail (in the author’s order): andersonvieira12@gmail.com; 17 priscilaalvesbezerra@gmail.com; anthonycarlos17@hotmail.com; 18 josieneveloso@yahoo.com.br; marcos.camara@ufrpe.br; vdoyle@agcenter.lsu.edu. 20 Corresponding author: Willie Anderson dos Santos Vieira 21 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 22 ABSTRACT 24 Colletotrichum is among the most important genera of fungal plant pathogens. 25 Molecular phylogenetic studies over the last decade have resulted in a much better 26 understanding of the evolutionary relationships and species boundaries within the 27 genus. There are now approximately 200 species accepted, most of which are 28 distributed among 13 species complexes. Given their prominence on agricultural 29 crops around the world, rapid identification of a large collection of Colletotrichum 30 isolates is routinely needed by plant pathologists, regulatory officials, and fungal 31 biologists. However, there is no agreement on the best molecular markers to 32 discriminate species in each species complex. Here we calculate the barcode gap 33 distance and intra/inter-specific distance overlap to evaluate each of the most 34 commonly applied molecular markers for their utility as a barcode for species 35 identification. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), histone-3 36 (HIS3), DNA lyase (APN2), intergenic spacer between DNA lyase and the mating- 37 type locus MAT1-2-1 (APN2/MAT-IGS), and intergenic spacer between GAPDH and 38 a hypothetical protein (GAP2-IGS) have the properties of good barcodes, whereas 39 sequences of actin (ACT), chitin synthase (CHS-1) and nuclear rDNA internal 40 transcribed spacers (nrITS) are not able to distinguish most species. Finally, we 41 assessed the utility of these markers for phylogenetic studies using phylogenetic 42 informativeness profiling, the genealogical sorting index (GSI), and Bayesian 43 concordance analyses (BCA). Although GAPDH, HIS3 and β-tubulin (TUB2) were 44 frequently among the best markers, there was not a single set of markers that were 45 best for all species complexes. Eliminating markers with low phylogenetic signal 46 tends to decrease uncertainty in the topology, regardless of species complex, and bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 47 leads to a larger proportion of markers that support each lineage in the Bayesian 48 concordance analyses. Finally, we reconstruct the phylogeny of each species 49 complex using a minimal set of phylogenetic markers with the strongest phylogenetic 50 signal and find the majority of species are strongly supported as monophyletic. 52 KEYWORDS 53 Accuracy 54 Anthracnose 55 Barcoding 56 Phylogenetic informativeness 57 Standardization 60 1. INTRODUCTION 61 Colletotrichum is among the largest groups of phytopathogenic fungi and 62 includes the causal agents of anthracnose and other diseases on seeds, stems, 63 leaves and fruits of important temperate and tropical crops (Cai et al., 2009, Cannon 64 et al., 2012). It is also among the most common genera of endophytic fungi, fungi 65 that live within plant organs without producing any symptoms of disease (Cannon et 66 al., 2012). Due to its economic and scientific importance, Colletotrichum was ranked 67 as the eighth most important phytopathogenic fungus in the world by plant 68 pathologists (Dean et al., 2012). 69 Species identification is necessary to understand disease epidemiology and 70 develop strategies to control the disease successfully (Cai et al., 2009). However, 71 Colletotrichum taxonomy and systematics has been a challenge since the genus was bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 72 introduced by Corda (1831). Colletotrichum species were historically circumscribed 73 on the basis of phenotypic features and a strong emphasis on the host species from 74 which the specimens were isolated under the assumption of host specificity, which 75 led to more than 900 species being recognized until revisionary work more than a 76 century after its introduction (von Arx, 1957; Sutton, 1980). Colletotrichum 77 identification based on morphological characters is problematic due to plasticity and 78 variation induced by experimental conditions (Vieira et al. 2017), and all life stages 79 are not frequently produced in culture (Samarakoon et al., 2018). The absence of 80 stable phenotypic characters has limited our understanding of phylogenetic 81 relationships within Colletotrichum and made the recognition of species boundaries 82 unreliable and confusing (Cai et al., 2009). To address this problem, Cai et al. (2009) 83 proposed a guideline for Colletotrichum species recognition based on a polyphasic 84 approach, which comprises the use of cultural, morphological, physiological and 85 pathogenicity characters in combination with phylogenetic analysis of nucleic acid 86 sequences. 87 The earliest phylogenetic studies of Colletotrichum using DNA sequences 88 were published by Mills et al. (1992) and Sreenivasaprasad et al. (1992). 89 Polymorphisms in the ITS1 region of the nrDNA were used to distinguish 90 Colletotrichum species. However, while the nrITS region is the most widely 91 sequenced region and has been chosen as the barcode locus for the Fungi, the utility 92 of this region is limited for systematic studies in Colletotrichum. Species diversity is 93 usually underestimated when based on nrITS sequences alone (Crouch et al., 94 2009a) and it has been demonstrated to have little phylogenetic utility (Doyle et al. 95 2013; Vieira et al. 2017). However, several additional markers have been applied for 96 multilocus phylogenetic inference to resolve the boundaries of cryptic species in the bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 97 genus (Cai et al. 2009; Damm et al. 2009; Doyle et al., 2013; Hyde et al., 2009; Lima 98 et al., 2013; Liu et al., 2016; Samarakoon et al., 2018; Veloso et al., 2018; Vieira et 99 al., 2014, 2017). 100 According to the most recent synopsis of the genus published in 2017, 188 101 Colletotrichum species have been described and incorporated into molecular 102 phylogenies. Among these species, 164 were distributed among 11 species 103 complexes and an additional 24 species had not been assigned to a species complex 104 (Marin-Felix et al., 2017). Additional species were recently described and three 105 additional clades were declared to represent new species complexes (Cao et al., 106 2018; Damm et al., 2019; Samarakoon et al., 2018). Due to its global distribution and 107 ecological and economic importance, research groups around the world are working 108 concomitantly to address regional diversity. However, the set of phylogenetic 109 markers used to discriminate species is variable by species complex and no standard 110 set of markers has been adopted based on objective criteria (Marin-Felix et al., 111 2017), making it difficult to combine data from disparate research groups and reliably 112 infer phylogenies and delimit species boundaries. Currently, thirteen different 113 molecular markers are commonly sequenced among the various Colletotrichum 114 species complexes: actin (ACT), DNA lyase (APN2), intergenic spacer between DNA 115 lyase and the mating-type locus MAT1-2-1 (APN2/MAT-IGS), calmodulin (CAL), 116 chitin synthase (CHS-1), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 117 intergenic spacer between GAPDH and a hypothetical protein (GAP2-IGS), 118 glutamine synthetase (GS), histone 3 (HIS3), nuclear rDNA internal transcribed 119 spacers (nrITS), mating type gene (MAT1-2-1), manganese-superoxide dismutase 120 (SOD2), and β-tubulin (TUB2). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 121 It is known that the efficiency of PCR amplification and the distribution of 122 phylogenetic informativeness of a given marker varies among species complexes 123 (Hyde et al., 2013). Most studies on the utility and reliability of individual markers 124 come from the Colletotrichum gloeosporioides complex (Cai et al., 2009; Liu et al., 125 2015; Sharma et al. 2013, Silva et al., 2012;2015; Vieira et al. 2017), and while the 126 C. gloeosporioides complex has been exhaustively studied in recent years, 127 recommendations on marker choice seem to be largely ignored. A classic case is the 128 utility of APN2/MAT-IGS, the most powerful marker to discriminate species within the 129 C. gloeosporioides complex: several recent studies described novel species within C. 130 gloeosporioides complex while excluding data from APN2/MAT-IGS (Costa et al., 131 2018; Diao et al. 2017; Fu et al., 2019; Jayawardena et al., 2016; Oliveira et al., 132 2018; Sharma et al., 2017; Silva et al., 2017; Sousa et al., 2018; Wang et al., 2019). 133 As mentioned above, this limits our ability to combine data from regional studies to 134 develop an accurate understanding of global diversity and phylogenetic relationships 135 within the genus. 136 While robust phylogenetic inference and reliable species delimitation relies on 137 the use of quality markers, studies on the performance of different molecular markers 138 for phylogenetic inference are missing for the majority of Colletotrichum species 139 complexes. In addition to the challenges discussed above, this also presents 140 practical problems for plant pathologists and ecologists who are looking to reliably 141 identify a large collection of isolates using molecular data. It is impractical for 142 researchers to sequence several loci, many of which may be of little phylogenetic 143 utility, across several hundred isolates simply for species identification. The aim of 144 the present study was to evaluate the phylogenetic informativeness of different 145 molecular markers used in Colletotrichum systematics and determine the optimal set bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 146 of markers for each species complex. From this, we hope to establish a consensus 147 on the minimal set of markers that can be used for Colletotrichum species 148 identification and delimitation and provide a practical reference for the large 149 community of researchers working on developing a better understanding of global 150 diversity, life history, and ecology of the genus. 152 2. MATERIAL AND METHODS 153 2.1 Datasets 154 We compiled several datasets to analyze the barcoding utility, phylogenetic 155 signal, and genealogical concordance of ACT, APN2, APN2/MAT-IGS, CAL, CHS-1, 156 GAPDH, GAP2-IGS, GS, HIS3, nrITS, and TUB2 for all Colletotrichum species 157 complexes described to date except for the C. caudatum complex, which was 158 excluded because nrITS was the only marker available for all species. These 159 datasets were compiled from published sequences retrieved from GenBank 160 (Supplementary File S1). Since testing the accuracy of prior species delimitations 161 were not the main focus of this study, we assumed that species boundaries 162 established in previous studies were accurate. 163 Thirteen species complexes were investigated in our study: Colletotrichum 164 acutatum, C. boninense, C. dematium, C. destructivum, C. dracaenophilum, C. 165 gigasporum, C. gloeosporioides, C. graminicola, C. magnum, C. orbiculare, C. 166 orchidearum, C. spaethianum and C. truncatum. Some species within each complex 167 were not included in the alignments due to the absence of sequences for several 168 markers since some of the analyses employed in the present work do not allow 169 missing data. Some markers were not analyzed due to a small number of species or 170 isolates with sequences available. The inclusion of these markers will drastically bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 171 reduce the number of species that can be included in each set of analyses (e.g. 172 GAPDH, HIS3, APN2 and APN2/MAT-IGS in the C. graminicola species complex). 174 2.2 Multiple sequence alignment 175 Multiple sequence alignments (MSA) of each locus were estimated individually 176 for each species complex. Sequences were compiled using the GenBank tool 177 implemented in MEGA 7 (Kumar et al., 2016). MSAs were estimated with the online 178 version of MAFFT 7 (Katoh et al., 2002; Katoh & Standley, 2013) using the G-INS-i 179 iterative refinement method and the 200PAM / k=2 nucleotide scoring matrix. 180 External gaps were trimmed in MEGA 7 before uploading MSAs to the GUIDANCE2 181 server (http://guidance.tau.ac.il/ver2/) (Sela et al., 2015) to access the alignment 182 confidence scores under the following parameters: MAFFT as the MSA algorithm; 183 max-iterate=0; pairwise alignment method=6mer; 100 bootstrap replicates. Unreliable 184 alignment regions were filtered by masking residues with scores below the lowest 185 cutoff, as proposed by Vieira et al. (2017). MSAs were converted to nexus format, 186 concatenated, and partitioned into a multilocus matrix using SequenceMatrix 1.8 187 (Vaidya et al., 2011). The number of invariable (I), variable (V), singletons (S) and 188 parsimony informative (PI) characters of the single locus alignments were calculated 189 using DnaSP 5.10 (Librado and Rozas,2009). 191 2.3 DNA barcoding 192 The effectiveness of markers to discriminate species within each species 193 complex was assessed by the barcode gap distance and intra/inter-specific distance 194 overlap (Hebert et al. 2003). Intra- and inter-specific distances were calculated for 195 each single locus alignment in MEGA 7. Single isolate species were removed from bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 196 alignments. Distances were calculated under the Kimura-2-parameter model, 197 allowing for substitution rates to differ among transitions and transversions, uniform 198 rates among sites, and gaps treated as pairwise deletions. Distance values were 199 sorted in Microsoft Excel Professional Plus 2016 and summary statistics were 200 calculated (maximum, minimum and mean distance). The barcode gap was 201 represented by the difference between the mean interspecific and intraspecific 202 distances (Hebert et al. 2003). The distance overlap percentage represents how 203 much the intraspecific distance overlaps with the interspecific distance and was 204 calculated as follows: max intraspecific distance ÷ max interspecific distance × 100. 205 Markers useful as barcodes will have a large barcode gap and a small intra/inter- 206 specific distance overlap. 208 2.4 Assessment of phylogenetic informativeness 209 The phylogenetic informativeness of markers commonly employed in 210 Colletotrichum systematics was estimated using the application PHYDESIGN 211 (Lopez-Giraldez and Townsend, 2011). Maximum likelihood (ML) trees were inferred 212 for each species complex using the concatenated alignments reduced to a single 213 representative isolate per species. Phylogenies were estimated in RAxML - HPC2 214 (Stawatakis, 2014) implemented on CIPRES Science Gateway portal 215 (https://www.phylo.org/portal2/home.action). ML tree searches were done assuming 216 the GTRGAMMA model and bootstrap support calculated with 1000 pseudoreplicates 217 (-m GTRGAMMA -p 12345 -k -f a -N 1000 -x 12345). ML trees were converted to 218 rooted ultrametric trees using the ‘chronos’ function in the ape package (Paradis et 219 al., 2004) using R Studio 1.1.442 (R Core Team, 2017). Trees were calibrated with 220 an arbitrary time scale with time=0 at the tips and time=1 at the root. The ultrametric bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 221 trees and the corresponding partitioned alignment were used as input files in 222 PHYDESIGN and the substitution rates were calculated using the program HyPhy 223 (Pond et al., 2005). The substitution rates estimated by the maximum likelihood 224 algorithm used by HyPhy are nonsensical for some sites in the alignment resulting in 225 very recent ‘phantom’ peaks that have no biological meaning. These peaks are likely 226 the result of indels or ambiguous sites in the alignment, therefore these alignment 227 positions with poorly estimated substitution rates were excluded from some genes 228 prior to phylogenetic informativeness profiling at the recommendation of the authors 229 of PhyDesign (http://phydesign.townsend.yale.edu/faq.html). Since the markers 230 included in this study and in most systematic studies of Colletotrichum do not require 231 more than two sequencing reads to sequence the full length of the locus 232 (representing the same sequencing effort and cost), the phylogenetic informativeness 233 values (PIV) were calculated on a net basis. The variable PImax represents the time 234 in which a given marker reaches the maximum PIV and was used to determine the 235 divergence time in which the marker is most informative (Fong and Fujita, 2011). We 236 ranked the markers according to the PIV values and the usefulness of markers was 237 assessed through the profile shape: low and flat curves represent the least 238 informative markers; high and sharp peaks represent the most informative markers. 239 The percentage of markers that resolve a given species was estimated using a 240 Bayesian Concordance Analysis – BCA (Ané et al., 2007; Larget et al., 2010). 241 Although BCA is a coalescent-based method to estimate species trees (Ané et al., 242 2007; Baum, 2007; Larget et al., 2010), this methodology can also be used to 243 quantify the proportion of markers that support a given clade, which is represented by 244 the concordance factor (CF). Individual locus trees were inferred in MrBayes 3.2.6 245 (Ronquist et al., 2012) implemented on the CIPRES cluster with four runs, each run bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 246 with four Markov chain Monte Carlo (MCMC) chains run for 10,000,000 generations, 247 sampled every 5,000 generations, totaling 2,001 trees per run. The frequency of 248 distinct topologies in the posterior distribution were summarized using mbsum 249 distributed with BUCKy 1.4.4 (Ané et al., 2007; Larget et al., 2010) skipping the first 250 25% of the trees as burn-in (-n 501). Primary concordance trees were estimated 251 using bucky and tree summary files output by mbsum were used as input files. 252 Concordance analyses were performed with a discordance factor (α) set at 1, four 253 MCMC chains, 1,000,000 generations, and the first 25% generations were discarded 254 as burn-in (-a 1 -k 4 -n 1000000 -c 4 -s1 23546 -s2 4564). 255 The Genealogical Sorting Index (GSI) was employed to identify the markers 256 that recover each species as monophyletic. GSI is an objective method that infers the 257 proportion of input trees for which a clade (species as applied here) are found to be 258 monophyletic and if the observed monophyly is greater than would be observed by 259 chance given the size of the data matrix (GSI=1 indicates monophyly) (Cummings et 260 al., 2008; Sakalidis et al., 2011), and can also be used to compare individual markers 261 according to their ability to discriminate species (Doyle et al., 2013). This 262 methodology can be applied to phylogenies inferred from a single locus as well as 263 multilocus analysis (Sakalidis et al., 2011). ML analyses were performed using the 264 single and multi-locus concatenated alignments. Species with a single isolate were 265 removed from the alignments. Analyses were carried out in RAxML as described 266 above with the number of pseudoreplicates reduced to 100 and outgroup isolates 267 specified prior to analysis. Rooted bootstrap trees were used as input files and each 268 tip was assigned to a species. GSIs were calculated using the GSI.py script and P- 269 value estimated from 100 permutations of each dataset (Cummings et al., 2008). The bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 270 GSI values were converted to heatmaps in the Heatmapper web server 271 (http://www.heatmapper.ca) to aid in visualization (Babicki et al., 2016). 273 2.5 Selection of best minimal sets of markers 274 The selection of a minimal set of optimal markers for robust phylogenetic 275 inference within each Colletotrichum species complex was based on results from the 276 phylogenetic informativeness profiling and GSIs. The markers were selected 277 according to the following criteria: 278 1) A minimum of three markers per complex based on ranking according to 279 PIV were selected. Three independent markers allow for the application of 280 the genealogical concordance phylogenetic species recognition criteria – 281 GCPSR (Dettman et al., 2003, Taylor et al. 2000), a commonly applied set 282 of criteria for phylogenetic species recognition in fungal systematics. 283 2) All species must be recognized as monophyletic by at least one of the 284 selected markers. GSIs were checked to confirm if each species in the 285 complex is recovered as monophyletic (GSI = 1) by at least one of the 286 selected markers. 287 3) ML trees were inferred from concatenated alignments of the three best 288 markers; if some species clade was poorly supported and/or all species 289 were not recovered as in the multilocus analyses with all markers 290 (unresolved relationships/polytomy), other markers were progressively 291 concatenated in decreasing order of phylogenetic informativeness until all 292 species were well resolved. 293 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 294 Once the best markers were chosen, the GSI was calculated for this 295 phylogeny to determine the level of species monophyly when only the best markers 296 are concatenated. The BCAs were also performed with this dataset to elucidate if the 297 species could be recognized by the majority of the selected markers. 299 3. RESULTS AND DISCUSSION 300 3.1 Alignment statistics 301 GAPDH, HIS3, and TUB2 were the most variable markers in the majority of 302 Colletotrichum species complexes, with PI characters ranging from 10—109, 11—82, 303 and 12—114, respectively (Table 1). PI characters for GS ranged from 63 to 93 and it 304 was the most variable marker within the C. gigasporum and C. orbiculare species 305 complexes. The APN2/MAT-IGS and GAP2-IGS, which are employed only in the C. 306 gloeosporioides species complex, had 192 and 115 PI characters, respectively, and 307 were the most variable markers within this complex. In contrast, nrITS presented the 308 fewest PI characters for most species complexes (0—36), followed by CHS-1 (3—45) 309 and ACT (3—63). 310 Most markers employed for Colletotrichum systematics comprise partial 311 sequences of orthologous protein-coding genes. These markers are composed of 312 introns flanked by long exons that are highly conserved. The GAPDH, GS, HIS3 and 313 TUB2 markers contain long intronic regions and APN2/MAT-IGS and GAP2-IGS 314 present long intergenic regions. This may explain why these markers are more 315 variable than the protein-coding loci. While protein-coding loci may be useful for 316 providing support along the backbone of the phylogeny within a species complex or 317 across the genus, markers with variable introns and intergenic sequences are 318 preferable for application at lower taxonomic levels (Schmitt et al., 2009). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 320 3.2 DNA barcoding feasibility 321 GAPDH had the largest barcode gap distance in seven of the 11 species 322 complexes evaluated. The percentage overlap between intra- and interspecific 323 distances was less than 20%, with the exception of the C. gigasporum and the C. 324 gloeosporioides species complexes (28.6% and 29.2%, respectively). GS has the 325 highest barcode distance with the lowest overlap for the C. gigasporum (0.12 and 326 12.1%) and C. orbiculare (0.08 and 2.7%) complexes. APN2/MAT-IGS had the 327 largest barcode gap distance (0.15) and the smallest overlap percentage (3.26%) 328 within the C. gloeosporioides species complex, making it the best candidate barcode 329 locus for the complex HIS3 had the largest barcode gap (0.026) within the C. 330 orchidearum complex with a relatively low overlap (16.9%), although CHS-1 had a 331 comparable barcode gap (0.025) with only a slightly higher overlap (21.6%). While 332 other markers, such as HIS3 and TUB2, are good candidate secondary barcode 333 markers in several complexes, nrITS was universally the poorest barcode candidate 334 with the lowest barcode gap distance within all species complexes. 335 Our results demonstrate that selecting a universal barcode marker for all 336 Colletotrichum species complexes among the markers currently being used is not 337 possible. An illustration of this is GAPDH. This marker is the best candidate barcode 338 marker for the majority of complexes, however it is among the worst barcode 339 candidates for the C. gloeosporioides species complex. While our results are not in 340 agreement with Cai et al. (2009), which carried out the first study evaluating the 341 markers to discriminate species within C. gloeosporioides species complex, only five 342 species were included in their analyses and the Musae and Kahawae clades (sensu 343 Weir et al. (2012)) were treated as single species. Moreover, while GAPDH was bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 344 chosen as the best marker relative to EF1α, ACT, CHS-1 and nrITS, the latter three 345 markers perform very poorly for species delimitation in the C. gloeosporioides 346 complex (Vieira et al., 2017) and, therefore, we did not include them in any of our 347 analyses. While GAPDH, with its large barcode gap and small overlap along with the 348 ease with which it can be amplified and sequenced, makes it among the best 349 barcode candidates across Colletotrichum as a whole, the selection of the best 350 barcode markers is dependent on the species complex. 351 The search for a universal barcode locus for the genus will require a 352 comparative analysis across the genomes of several species in all species 353 complexes. Intergenic sequences in syntenic regions of the genome are good 354 candidates if APN2/MAT-IGS and GAP-IGS are any indication. Intergenic sequences 355 may provide fast-evolving phylogenetic markers to be used for population genetic 356 and phylogenetic studies on fungal species complexes (Magain et al., 2017). 358 3.3 Optimal markers for phylogenetic inference 359 The most informative markers differ among the species complexes within 360 Colletotrichum. Net phylogenetic informativeness profiles and their respective 361 ultrametric trees are presented in Fig. 2. PIV and PImax values are summarized in 362 Table 1. The GSI analyses illustrate the ability of a given marker to recover species 363 as monophyletic is dependent on the species complex (Fig. 3, GSI values presented 364 in Supplementary File S2). Most species that were monophyletic in the multilocus 365 tree with all markers (GSI near to 1) were also recovered as monophyletic when only 366 the most informative markers were concatenated. In parallel, the BCAs revealed that 367 the proportion of markers supporting individual species-level clades (expressed as bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 368 concordance factors) increase when the less informative markers are removed from 369 the analyses. 370 Since the suite of markers differ by species complex and the performance of a 371 given marker differs among the species complexes, the results of the phylogenetic 372 informativeness profiling, GSI, Bayesian concordance analyses and selection of the 373 minimum set of markers for robust phylogenetic inference are summarized below for 374 each species complex. 376 3.3.1 Colletotrichum acutatum s. l. 377 HIS3, GAPDH and TUB2 are the most phylogenetically informative markers 378 within the C. acutatum species complex, with PImax at 0.67, 0.78 and 0.99, 379 respectively (Fig. 2). All markers currently used for systematics of the C. acutatum 380 complex have an optimum inferential timescale varying from 0.67 to 0.99, which is 381 useful to resolve deeper relationships. However, the absence of markers with lower 382 PImax negatively impacts the our ability to infer relationships among recently 383 diverged species. This is evident by the low internal node support within Clade 1 and 384 Clade 2 sensu Damm et al. (2012a), which are the most recently diverged lineages 385 within the complex (Supplementary File S3). Species are strongly supported in the 386 concatenated multilocus analysis (ML support ≥ 70%, Supplementary File S3), 387 whereas the relationship among them are poorly resolved even when only the best 388 markers were used to build the tree. In contrast, nodes throughout Clade 5 are 389 strongly supported. The inclusion of a marker with PImax near to 0.2 could improve 390 the support on the deeper nodes of the Clades 1 and 2. 391 Most species in C. acutatum s. l. could be resolved by the majority of the 392 markers only when the optimal markers are combined (Fig. 4B). Seven (C. austral, C. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 393 chrysanthemi, C. fioriniae, C. johnstonii, C. lupini, C. nymphaeae and C. tamarilloi) 394 among the 17 multiple-isolate species were supported by the majority of individual 395 genes (CF≥0.66) in the BCA when all six loci were used in the analysis (Fig. 4A), 396 which means that four out of six markers support the monophyly of these species. 397 The CF increases when the analyses included only HIS3, GAPDH and TUB2, and 398 only four (C. cosmi, C. costaricense, C. paranaense and C. phormii) out of 17 399 multiple-isolate species presented CF<0.66 (Fig. 4B). The GSI (Fig. 3) indicates that 400 C. melonis is recovered by GAPDH and C. phormii by HIS3. The GSI for 401 Colletotrichum costaricense was nearly 1 for both GAPDH and TUB2 (0.95 and 0.99, 402 respectively) and less than 0.5 for all other markers. This is consistent with a CF of 403 0.63, indicating that nearly 2 of the 3 loci included in the analysis support the 404 monophyly of the species. Colletotrichum paranaense had a low GSI for most 405 markers and CF=0.22, which means that no individual marker fully supports this 406 species as monophyletic. However, C. paranaense is strongly supported in the 407 multilocus concatenated analysis and has a high GSI when analyses included all or 408 the best markers. These results clearly show that the multilocus trees masks the 409 incongruences among the individual gene trees or the use of markers with low 410 phylogenetic signal. 411 Colletotrichum paranaense was described by Bragança et al. (2016) as a 412 species closely related to C. limetiicola and C. melonis (Damm et al., 2012a). 413 Phylogenetic species recognition was not employed in the delimitation of C. 414 paranaense. The species was recognized based on the topology of the multilocus 415 tree and differed from C. limetticola and C. melonis by percentage identity according 416 to blastn searches. Our results suggest that the relationship among these species 417 and the delimitation of C. paranaense needs to be revisited. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 419 3.3.2 Colletotrichum boninense s. l. 420 CAL, TUB2 and GAPDH were the most informative markers to resolve species 421 within C. boninense s. l. (Fig. 2). All three markers are comparable in levels of 422 phylogenetic informativeness, with PIV ranging from 47 to 53. However, these 423 markers differ with respect to the optimal timescale at which they are informative. 424 CAL and TUB2 peaked at 0.58 and 0.63, respectively, providing more signal towards 425 the root of the phylogeny. In contrast, GAPDH reached maximum informativeness at 426 0.24, which is more useful for resolving species in this complex, since the divergence 427 epoch for most species is near 0.2. 428 Most species can be resolved by most of the markers according to the 429 concordance analyses that include all markers and only the best markers (CF>0.57 430 and >0.66, respectively) (Fig. 5B). GAPDH supports most species as monophyletic 431 (GSI 0.98—1) (Fig. 2), with the exception of three species (C. catinaense, C. 432 constrictum and C. limonicola). According to Damm et al. (2012b), all the species 433 within the C. boninense species complex can be identified by sequencing GAPDH. In 434 contrast, our results show that C. catinaenese, C. constrictum and C. limonicola are 435 weakly supported as monophyletic by GAPDH (GSI = 0.51, 0.69 and 0.59, 436 respectively) and could only be recovered as monophyletic in analyses of TUB2, HIS 437 or ACT. Colletotrichum cymbidiicola and C. novae-zelandiae were recovered as 438 monophyletic with GAPDH and C. limonicola by ACT and TUB2 (Fig. 3). Thus, 439 concordance factors for these species remained low even when the best markers 440 were used in the BCA. Additionally, C. novae-zelandiae was the only species for 441 which GSI<1 in both multilocus analyses (Fig. 3). Colletotrichum catinaense could be 442 resolved by two of the best markers (CAL and TUB) and by ACT, which led the CF to bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 443 shift from 0.56 in the BCA with all markers to 0.73 when only the best markers were 444 included (Fig. 5). The use of CAL, TUB2 and GAPDH leads to a slight increase in 445 support across the multilocus ML tree when compared with the tree reconstructed 446 with the whole dataset (Supplementary File S4), with the exception of Colletotrichum 447 novae-zelandiae, which remained poorly supported. 449 3.3.3 Colletotrichum dematium s. l. 450 GAPDH, HIS3 and TUB2 presented high and sharp profiles (Fig. 2), with 451 PImax 0.05—0.06 and were the most useful markers to discriminate species within 452 the C. dematium species complex. HIS3 was the only marker able to discriminate all 453 species in the complex. nrITS, CHS-1 and ACT presented low and flat curves and 454 were the least informative markers (Fig. 2). Colletotrichum anthrisci and C. circinans 455 could be resolved by all markers (CF=1, Fig. 6), while Colletotrichum spinaceae was 456 resolved as monophyletic by all markers except nrITS. The exclusion of nrITS in the 457 BCA increased the CF from 0.86 to 1. The relationship between Colletotrichum 458 dematium and C. lineola was not clearly resolved in the multilocus analyses. 459 Although C. dematium isolates were well supported as monophyletic in the tree 460 inferred from all markers, C. lineola isolates remained paraphyletic (Supplementary 461 File S5). When only GAPDH, HIS3 and TUB2 were considered, the isolates of C. 462 dematium and C. lineola were placed together in a poorly supported clade. 463 Colletotrichum lineola was recovered as monophyletic only by HIS3 (GSI=1). Both 464 ACT and GAPDH could also resolve this species, albeit with low GSI (0.83 and 0.73 465 respectively). In contrast, these three markers recover C. dematium with high level of 466 monophyly (GSI 0.97—1). Colletotrichum lineola presented low monophyly level in all bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 467 multilocus GSI analysis and C. dematium was not monophyletic when only the best 468 markers were considered. 469 In the Damm et al. (2009) study, the isolates of C. dematium and C. lineola are 470 separated into two short-branch clades. They chose to retain both taxa, but raised 471 the hypothesis that C. dematium and C. lineola are different populations within the 472 same species. We performed the GSI analyses with the multilocus trees combining 473 C. dematium and C. lineola in the same group (data not shown). The group was 474 recovered as monophyletic when all markers and only the best markers are 475 concatenated (GSI=1 and 0.99 respectively), which support the hypothesis that C. 476 dematium and C. lineola are the same species. Colletotrichum eryngiicola, C. 477 hemerocallidis, C. insertae, C. quiquefoliae and C. sonchicola were placed together 478 with C. dematium and C. lineola in a polytomous clade in Samarakoon et al. (2018) 479 study, which indicates that those species may also be members of the group C. 480 dematium/ C. lineola. These species were not included in our analyses due to 481 absence of HIS3 and TUB2 sequences. In the future, species boundaries within this 482 lineage need to be revisited. 484 3.3.4 Colletotrichum destructivum s. l. 485 HIS3, TUB2 and GAPDH were the most informative markers to resolve 486 species within C. destructivum complex (Fig. 2). GAPDH and HIS3 possess the 487 phylogenetic signal to resolve shallow clades (PImax=0.28 and 0.37 respectively), 488 whereas TUB2 performs better in the deep branches (PImax=0.73). HIS3 was the 489 most informative marker and was able to six (C. destructivum, C. lentis, C. lini, C. 490 tabacum, C. utrechtense and C. vignae) out of the nine species analyzed as 491 monophyletic (GSI 0.98—1) (Fig. 3). Although GAPDH is one of the best markers, bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 492 only four species (C. fuscum, C. lentis, C. tabacum and C. vignae) were highly 493 monophyletic in the topologies provided by this locus (GSI 0.9—1) (Fig. 3). 494 Colletotrichum americae-borealis was not monophyletic in the topology of any single 495 or multi-locus trees (GSI 0.34—0.89). Four (C. destructivum, C. lentis, C. tabacum, 496 C. utrechtense) out of nine multiple-isolates species could be recovered by most of 497 the genes in the concordance tree (CF≥0.66) inferred from all markers (Fig. 7A). 498 These same four species plus C. vignae were recovered by most of the genes 499 (CF≥0.66) when only the optimal markers were considered. Moreover, the CF and 500 the bootstrap supports of the internal nodes increased when only the best markers 501 were combined (Fig. 7 and Supplementary File S6). 502 The clade comprised by C. americae-borealis and C. lini forms a polytomy 503 when only the three best markers were used (Supplementary File S6). We also 504 tested the inclusion of ACT or nrITS in the concatenated dataset. However, this clade 505 remains a polytomy and can only be resolved when all markers are concatenated. 506 The overall GSI values also reduced when ACT or nrITS was included in the 507 multilocus analysis (data not shown). Thus, it is not reasonable to include ACT and 508 nrITS in the analysis due to their low phylogenetic informativeness. Other markers 509 with greater phylogenetic signal need to be tested for this complex in order to resolve 510 the relationships among the unresolved clades and clarify the species identities. 512 3.3.5 Colletotrichum dracaenophilum s. l. 513 TUB2, GAPDH and HIS3 presented the highest PIV (58, 52 and 36 514 respectively) and were the most informative markers to distinguish species within the 515 C. dracaenophilum complex (Fig. 2). All the markers peaked around the same 516 timescale (PImax 0.39—0.46) and provide robust support for relationships in both bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 517 deep and shallow nodes. Although TUB2, GAPDH and HIS3 were the most 518 informative markers, all the markers are informative enough to discriminate species 519 in C. dracaenophilum s. l. (GSI 0.99—1) (Fig. 3). This is corroborated by the high 520 concordance factors (CF=1) in BCAs (Fig. 8) and 100% support in multilocus 521 analyses (Supplementary File S7) when all or the best markers were combined. 523 3.3.6 Colletotrichum gigasporum s. l. 524 GS, CAL and GAPDH were the most powerful markers to discriminate species 525 within C. gigasporum s. l. (Fig. 2), with PImax at 0.1, 0.44 and 0.28 respectively. 526 Although GS, CAL and GAPDH were the most informative markers, all the markers 527 and the concatenated datasets were able to discriminate all species within the C. 528 gigasporum complex (GSI 0.99—1) (Fig. 3), as reported by Liu et al. (2014). All 529 species presented maximum CF (Fig. 9) and ML support (Supplmentary File S8) 530 independently of which set of markers was included in these analyses. 532 3.3.7 Colletotrichum gloeosporioides s. l. 533 APN2/MAT-IGS, GAP2-IGS and APN2 were the most informative markers to 534 separate species within the C. gloeosporioides species complex (Fig. 2). The 535 informativeness profiles indicate APN2/MAT-IGS and APN2 are of peak 536 informativeness at 0.19 and 0.17, respectively, and are informative for shallow 537 divergences, whereas GAP2-IGS has an optimal timescale at 0.42 and provides 538 more signal for resolving deep nodes. This range in values for peak informativeness 539 led to high support at both deep and shallow nodes when only the most informative 540 markers were used for phylogenetic inference (Supplementaray File S9). APN2/MAT- 541 IGS could separate all the species included in the analyses, although only C. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 542 fragariae, C. queenslandicum, C. siamense and C. tropicale reached a GSI of 1 (Fig. 543 3). Moreover, all species were recovered as monophyletic in all multilocus analyses. 544 Colletotrichum siamense was the only species that could not be recovered by most of 545 the markers, and its CF reduced from 0.59 in all-markers analysis to 0.38 in the best- 546 markers analysis (Fig. 10B). The CF reduced due to removing CAL and TUB2, which 547 also recover C. siamense as monophyletic (GSI=0.99). 548 ACT, CHS-1 and nrITS were not included in our analyses because these 549 markers were previously reported as the worst markers to distinguish species in C. 550 gloeosporioides complex (Vieira et al., 2017). In the present study, rather than 551 restating the results of Vieira et al. (2017), we evaluated if the seven markers 552 proposed by those authors are needed for diversity studies and species assignment. 553 We determined sequences of APN2/MAT-IGS, GAP2-IGS and APN2 are sufficient to 554 resolve the species in C. gloeosporioides s. l.. However, sequences of GAP2-IGS 555 and APN2 are not available for all species within C. gloeosporioides s.l.. These data 556 should be generated for all species within the species complex. 557 Colletotrichum gloeosporioides s. l. is the most studied species complex in the 558 genus. More than 10 different markers have been used for systematics and 559 taxonomy of the C. gloeosporioides complex over the last 10 years. However, there 560 is no agreement about which markers should be used for species recognition. The 561 first attempt to determine the best markers was done by Cai et al. (2009), in which 562 GAPDH was the best marker to discriminate species in C. gloeosporioides s. l., and 563 the set composed by ACT, GAPDH, nrITS and TUB2 was recommended for 564 multilocus analysis. Weir et al. (2012) highlighted that although GAPDH is one of the 565 most effective markers to distinguish species within C. gloeosporioides complex, the 566 combination with GS is necessary to distinguish some species. Our study shows that bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 567 GAPDH was the least informative marker among the ones included in this study for 568 C. gloeosporioides s. l.. This marker was the least variable (Table 1), and had the 569 smallest barcode gap and the largest overlap distance (Fig. 1). Colletotrichum 570 fragariae was the only species recovered as monophyletic (GSI=1) by this marker 571 (Fig 2). Based on this, GAPDH is considered one of the worst barcode candidates for 572 the C. gloeosporioides complex among the markers tested in the present study. 573 More recently, several studies demonstrate the singular ability of APN2/MAT- 574 IGS to discriminate species in the C. gloeosporioides complex (Sharma et al., 2013, 575 2015; Vieira et al., 2014), although others previously demonstrated the utility of this 576 marker (Du et al., 2005; Rojas et al., 2012; Silva et al., 2012). The main issue in 577 using APN2/MAT-IGS was the splitting out of C. siamense in a species complex, in 578 which several monophyletic lineages could be revealed by the phylogeny inferred by 579 this marker (Sharma et al., 2013). Although the multilocus analysis had been done, 580 the lineages identity was confirmed only based on the APN2/MAT-IGS, since this 581 marker performs good as well as the multilocus matrix. Other studies use the same 582 criteria to discriminate species and several species within C. siamense s. l. were 583 described (Sharma et al., 2015; Vieira et al., 2014). Later, Liu et al. (2016) use 584 coalescent methods for phylogenetic species delimitation and synonymize all those 585 species in the complex into C. siamense. It was revealed incongruences among the 586 APN2/MAT-IGS tree and other individual gene trees. The study clarifies how 587 multilocus analyses can mask discordances among individual gene trees and lead to 588 species misidentification. 589 The combination of APN2/MAT-IGS and GS were proposed as the barcode to 590 delimit species within C. gloeosporioides complex (Liu et al., 2015). These two 591 markers were collectively powerful enough to discriminate the 22 species included in bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 592 the study and produced the same topology as that inferred from a 6 marker 593 multilocus dataset (ACT, CAL, GAPDH, GS, nrITS and TUB2). We tested the 594 combination of APN2/MAT-IGS with each remaining marker individually and all 595 resulting trees were similar in topology (data not shown). Some species, such as C. 596 siamense, are polyphyletic in our GAPDH and GS trees, which belies the 597 incongruence with the multilocus and the other individual gene trees as presented by 598 Silva et al. (2012) and Vieira et al. (2017). Mating-type associated markers, such as 599 APN2/MAT-IGS and MAT1-2, had fast evolutionary rates and high variability, and 600 may dominate the topology of multilocus trees (Liu et al., 2016). Since the 601 combination of APN2/MAT-IGS with any other marker produces similar topologies, 602 the combination with any other marker besides GAPDH and GS is preferable in order 603 to avoid inconsistencies in species delimitation within the C. gloeosporioides 604 complex. While we find that APN2 is another informative marker that should be 605 incorporated into diversity studies of the C. gloeosporioides complex, the proximity of 606 this marker to the APN2/MAT-IGS region suggests it is part of the same linkage 607 group and thus may not represent an independent sample of an organisms 608 evolutionary history. If this is the case, linking substitution models and inferred trees 609 across these two loci would be required and the addition of another locus may be 610 necessary to fulfill the expectations under GCPSR. 611 Although the majority of Colletotrichum species belong to the C. 612 gloeosporioides complex, identification and description of taxa within this complex is 613 the most problematic. Sequences of the best markers, mainly APN2 and GAP2-IGS, 614 have not been sequenced for the majority of species, which prohibits the detection of 615 these species using these markers. A priority should be to generate sequences of the 616 best markers for all species to avoid further misidentification and the introduction of bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 617 dubious taxa. Until this happens, the description of new species within the C. 618 gloeosporioides complex will likely require sequencing multiple additional markers to 619 be certain of their novelty. 621 3.3.8 Colletotrichum graminicola s. l. 622 TUB2 was the most informative marker, followed by nrITS and ACT (Fig. 2). In 623 contrast, CHS-1 was ranked as the worst marker for the C. graminicola species 624 complex. GSI and BCA analyses could not be performed for this complex because 625 only one isolate per species could be included in the analyses. Other markers such 626 as APN2, APN2/MAT-IGS, GAPDH, HIS3, MAT1-2 and SOD2 were also used to 627 identify species within C. graminicola complex (Cannon et al., 2012; Crouch et al. 628 2009b, c; Crouch and Tomaso-Peterson, 2012; Du et al., 2005; Moriwaki and 629 Tsukiboshi, 2009; O'Connell et al., 2012; Tao et al., 2013). However, sequences for 630 these markers are missing for several species. According to our results for other 631 complexes, APN2, APN2/MAT-IGS, GAPDH and HIS3 are likely to be more powerful 632 markers than those that we could include in the analyses for the C. graminicola 633 complex. Sequences for these markers need to be generated for isolates of several 634 species in order to establish a better set of optimal markers for phylogenetic 635 inference and species discrimination in the C. graminicola complex. 637 3.3.9 Colletotrichum magnum s. l. 638 HIS3, TUB2 and GAPDH were the most informative markers to discriminate 639 species within the C. magnum complex (Fig. 2). Although HIS3 presented the highest 640 PIV (26), TUB2 presented a higher PImax (0.52 versus 0.44) and is able to separate 641 more species than HIS3. Only C. brevisporum and C. magnum were accounted for bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 642 by more than one isolate and could be analyzed in the BCA and GSI analyses (Fig. 643 3). Colletotrichum brevisporum reach high levels of monophyly only in analyses of 644 the concatenated dataset (GSI 0.98—1, Fig. 3) and only one gene can fully resolve 645 this clade (CF=0.4, Fig 11). In contrast, C. magnum is recovered as monophyletic by 646 the multilocus dataset and by the best markers. However, the CF reached 1 only 647 when the best markers were concatenated (Fig 11). Colletotrichum magnum 648 remained highly supported when only the best markers were used to reconstruct the 649 phylogeny of the C. magnum complex (Supplementary File S10B) with only a slight 650 decrease in support at some internal nodes. The species C. liaoningense was not 651 included in the analyses due to sequence deposition errors detected by Damm et al. 652 (2019) from the study were C. liaoningense was described (Diao et al., 2017). Damm 653 et al. (2019) concluded that C. liaoningense needs to be revisited. 655 3.3.10 Colletotrichum orbiculare s. l. 656 GS stood out from all markers as the most informative (PIV=69) in the C. 657 orbiculare complex, followed by HIS3 and GAPDH (PIV=37 and 32 respectively) (Fig. 658 2). GS peaks at 0.53 and can discriminate the majority of species and provide robust 659 support for the relationships among them, since the divergence time for most species 660 is about 0.3. On the other hand, HIS3 and GAPDH are useful to discriminate some 661 recently diverged species (PImax=0.13 and 0.17 respectively) such as C. orbiculare 662 × C. sidae and C. trifolii × C. malvarum. GS was the marker that recovered more 663 species as monophyletic (GSI=1), with the exception of C. sidae (GSI=0.89) (Fig. 3). 664 Colletotrichum lindemuthianum was the only species that was not recovered as 665 monophyletic when all or the best markers were concatenated (both GSI=0.78), 666 which is likely due to the variability within this species as currently circumscribed. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 667 Damm et al. (2013) split C. lindemutianum into two different lineages (C. 668 lindemuthianum 1 and 2), which are observed in the nrITS and GS trees. In our 669 multilocus trees, the isolates CBS133.57 and CBS131.57 where moved to the clade 670 C. lindemuthianum 1 and CBS150.28 to the clade C. lindemuthianum 2. This result is 671 discordant with that in Damm et al. (2013) and we conclude that the terminal clades 672 within C. lindemuthianum represent intraspecific variability. Thus, both C. 673 lindemuthianum lineages were not considered separate species in our analysis. The 674 CF of both deep and shallow nodes increases significantly when only the best 675 markers were analyzed (Fig.12). Reducing the set of markers does not cause 676 significant differences in the tree topology or clade support (Supplementary file S10). 678 3.3.11 Colletotrichum orchidearum s. l. 679 TUB2 and HIS3 were the most powerful genes capable of discriminating 680 species within the C. orchidearum complex (PIV=17 and 16 respectively), followed by 681 CHS-1 (PIV=12) (Fig. 2). TUB2 and HIS3 are good markers to discriminate all 682 species in the complex and support the relationships among them (PImax=0.8 and 683 0.67 respectively), whereas CHS-1 can help to distinguish and support recently 684 diverged species (PImax=0.47). All species were recovered with high levels of 685 monophyly with TUB2 (GSI=0.99—1), and also by both multilocus analyses 686 (GSI=0.97—1) (Fig. 3). Colletotrichum musicola was the only species that could be 687 recovered by all markers evaluated (CF=1) (Fig. 13). The CFs increased when the 688 less informative markers were removed from the analysis and most species were 689 recovered by all markers (CF=1), with the exception of C. cliviicola and C. plurivorum 690 (CF≥0.66). Damm et al. (2019) also reported that some clades were not supported by 691 some of the individual gene analyses. All deep and shallow nodes retained high bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 692 support when only the best markers were combined in our analyses (Supplementary 693 File S12B). 695 3.3.12 Colletotrichum spaethianum s. l. 696 GAPDH, TUB2 and ACT were the best markers to separate species within the 697 C. spaethianum complex (Fig. 2). These markers peak at approximately 0.4 (PImax 698 0.47—0.77), which is in the range of where most species in the complex diverge 699 indicating they can separate the majority of species in this complex. However, 700 GADPH is the only marker that recovers all species as monophyletic (GSI 0.98—1) 701 (Fig. 3). All species reached complete monophyly (GSI=1) in both multilocus trees. 702 Colletotrichum liriopes and C. spaethianum could be recovered by most of the genes 703 only when the best markers are considered in the BCA (CF=0.77 and 0.70, 704 respectively) (Fig. 14B). All species remain strongly supported when only the best 705 markers were concatenated (Supplementary Figure S13B). 706 HIS3 appears to be a good marker in several other Colletotrichum complexes 707 according to the present study, and may be among the three best markers for C. 708 spaethianum complex. However, sequences of HIS3 and other markers such as 709 CHS-1 and CAL are available only for one isolate of several species and we cannot 710 include them in our analyses. Sequences of these markers need to be generated for 711 other isolates to determine if one of these markers could be used to substitute for 712 ACT, which does not perform well for separating Colletotrichum species. 714 3.3.13 Colletotrichum truncatum s. l. 715 GAPDH was clearly the most informative marker (PIV=57), followed by TUB2 716 and ACT (PIV=33 and 23 respectively) (Fig. 2). All markers peaked above 0.4 and bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 717 were able to discriminate most species. Colletotrichum acidae and C. curcumae were 718 monophyletic (GSI=1) in all single and multilocus datasets (Fig. 3) with CFs equal to 719 1 in analyses of both all and best markers BCA (Fig. 15). These species were also 720 supported by maximum bootstrap values in the multilocus trees (Supplementary File 721 S14). Colletotrichum corchorum-capsularis and C. truncatum were not recovered as 722 monophyletic by any dataset. Moreover, C. truncatum is paraphyletic or polyphyletic 723 in the multilocus trees (Fig. 15, Supplementary File S14), which leads us to the 724 conclusion that C. corchorum-capsularis, as circumscribed by Niu et al. (2016), 725 cannot be recognized as a species distinct from C. truncatum. Our results strongly 726 suggest that C. corchorum-capsularis and C. truncatum may be the same species 727 and isolates of both species were placed together in a clade with high CF (≥0.96) in 728 the BCAs and maximum support in the multilocus analyses (Fig. 15, Supplementary 729 File S14). Additional work is needed using the best markers for the C. truncatum 730 complex and objective species recognition methods to determine the taxonomic 731 status species boundaries of C. corchorum-capsularis and C. truncatum. 733 4. Conclusions 734 We used phylogenetic informativeness profiling, maximum likelihood and 735 coalescent-based phylogenetic analyses, measures of barcode utility, and 736 genealogical sorting indices to assess the performance of the several molecular 737 markers used in Colletotrichum systematics and taxonomy across all known species 738 complexes. While HIS3, GAPDH, and TUB2 were among the best markers for most 739 of the complexes, the optimal set of markers is not always the same across all 740 complexes. ACT, CHS-1 and nrITS were the worst markers and, as previously 741 proposed for the C. gloeosporioides complex (Vieira et al., 2017), they can be bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 742 discarded from the phylogenetic analysis of almost all species complexes. ACT was 743 retained in the set of best markers for the C. graminicola, C. spaethianum and C. 744 truncatum complexes to achieve a minimum of three markers as proposed in the 745 methodology of the present study. However, few markers were included for these 746 complexes due to missing data, therefore additional markers need to be sequenced 747 and their performance evaluated. The analyses of C. caudatum, C. dematium, C. 748 graminicola and C. spaethianum complexes were the most impacted by the 749 excessive amount of missing data for the majority of the markers, which highlights 750 the importance of selecting a standard set of markers to delimit species. Similarly, 751 several isolates and/or species were not included in the marker analyses for C. 752 gloeosporioides s.l. due to selective data acquisition by different research groups. It 753 is not clear how the inclusion of sequences from these isolates might impact our 754 results. Sequences need to be generated for these markers and/or species to 755 provide more decisive results. We have also identified species complexes, which will 756 need to be revisited in the future, in which it appears species have been misidentified 757 (C. acutatum, C. dematium and C. truncatum). 758 Selecting the optimal markers to sequence for biodiversity studies on 759 Colletotrichum will impact Colletotrichum studies in a few ways. First, species 760 recognition will likely be more accurate and robust by avoiding the confounding effect 761 of including markers with low phylogenetic signal in the analyses. Secondly, 762 phylogenetic studies of Colletotrichum will become more economical, since 763 sequencing markers with low informativeness represents a low return on investment. 764 Finally, if research groups take guidance from this study, we are more likely to see a 765 consensus developed on the data acquired for phylogenetic studies on Colleotrichum 766 and we will be closer to a global assessment with combinable data. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 767 Researchers around the world continue to have an interest in documenting the 768 diversity of Colletotrichum species associated with economically important plant 769 species. However, this work is labor intensive and expensive because several 770 hundred isolates are typically screened and a paucity of distinctive morphological 771 characters necessitates DNA sequencing. The expense has been unnecessarily 772 compounded by the lack of an objective and comprehensive assessment of the utility 773 of existing markers for phylogenetic inference and species identification/delimitation 774 and the lack of a consensus on the markers to be used. We hope the results 775 presented here will help to address this problem. While the optimal markers differ by 776 species complex, our results provide some guidance on the most efficient path to 777 document and describe diversity within Colletotrichum. Our results also show that for 778 the accurate identification and delimitation of Colletotrichum species, a small set of 779 markers with strong phylogenetic signal is more suitable than a large set including 780 markers with both weak and strong phylogenetic signal. GAPDH is among the 781 optimal set of markers for 10 of the 13 species complexes in Colletotrichum, followed 782 by TUB2 (10 of 13), and HIS3 (7 of 13). Therefore, GAPDH is a good marker to 783 sequence for initial diversity screening and assigning isolates to a species complex 784 because data for this marker is available for the majority of species within the genus. 785 However, selection of additional markers for phylogenetic inference and species 786 delimitation will depend on the species complex. 787 Finally, while we recommend the optimal markers for species recognition 788 within Colletotrichum in order to improve diversity studies in the genus, our 789 understanding of evolutionary relationships among species remains poorly resolved. 790 Improving our understanding of relationships among taxa within Colletotrichum will 791 require more robust genomic sampling. Genome sequencing is underway for many bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 792 species of Colletotrichum, however a comprehensive phylogenomic study of the 793 genus is needed. 795 Acknowledgments 796 This study was funded by “Conselho Nacional de Desenvolvimento Científico 797 e Tecnológico – CNPq” (Universal number 408724/2018-8). Willie A. S. Vieira 798 acknowledges the “Fundação de Amparo à Ciência e Tecnologia do Estado de 799 Pernambuco – FACEPE” for the Postdoc fellowship (Process number BFP-0040- 800 5.01/16). Marcos P. S. Câmara acknowledge CNPq for the research productivity 801 fellows. Josiene S. Veloso acknowledge the “Coordenação de Aperfeiçoamento 802 Pessoal de Ensino Superior – CAPES” and the “Programa Nacional de Pós- 803 Doutorado/CAPES – PNPD/CAPES” for the Posdoc fellowship. This work represents 804 partial fulfillment of the degree of Master of Phytopathology for Priscila A. Bezerra. V. 805 P. Doyle would like to acknowledge funding support from the Louisiana Board of 806 Regents Research Competitiveness Subprogram (LEQSF(2016-19)-RD-A-01). 808 Glossary 809 Appressorium: specialized cell produced by some phytopathogenic fungi which is 810 used to infect plant hosts. 811 Conidium: asexual spore of Ascomycota and Basidiomycota. 812 Endophytic fungi: fungi that grow inside the plant tissues without causing disease 813 symptoms. 814 Phytopathogenic fungi: fungi that cause plant diseases. 815 Sensu lato (s. l.): taxonomic terminology used to reference species complexes (C. 816 acutatum species complex = C. acutatum s. l.). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 817 Sensu stricto (s. s.): when is necessary to refer the species with the same name of 818 the complex (C. acutatum s. s. is a species within C. acutatum s. l.). 819 Species complex: major clades strongly supported within Colletotrichum genus tree. 820 These clades include phylogenetic species closely related which most are 821 indistinguishable based on phenotypical characters (e.g. conidial and appressorial 822 shape and size, growth rate, color of colonies). Species complexes get the same 823 name of the species within them that is more known or that was firstly described. 824 In some cases, members within a given species complex share peculiar conidial 825 characteristics: C. acutatum – conidia with acute ends; C. boninense – presence of a 826 prominent scar (hilum) at the base of the conidium; C. caudatum – conidia with a 827 filiform appendage at the apex; C. gigasporum – longest and widest conidia within 828 the genus. 830 References 831 Ané, C., Larget, B., Baum, D.A., Smith, S.D., Rokas, A., 2007. Bayesian estimation 832 of concordance among gene trees. Molecular Biology and Evolution 24, 412– 833 426. 834 Babicki, S., Arndt, D., Marcu, A., Liang, Y., Grant, J.R., Maciejewski, A., Wishart, 835 D.S., 2016. Heatmapper: web-enabled heat mapping for all. Nucleic Acids 836 Research 44, W147–W153. 837 Baum, D.A., 2007. Concordance trees, concordance factors, and the exploration of 838 reticulate genealogy. Taxon 56, 417–426. 839 Bragança, C.A.D., Damm, U., Baroncelli, R., Massola Júnior, N.S., Crous, P.W., 840 2016. Species of the Colletotrichum acutatum complex associated with 841 anthracnose diseases of fruit in Brazil. Fungal Biology 120, 547–561. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 842 Cai, L., Hyde, K.D., Taylor, P.W.J., Weir, B.S., Waller, J., Abang, M.M., Zhang, J.Z., 843 Yang, Y.L., Phoulivong, S., Liu, Z.Y., Prihastuti, H., Shivas, R.G., McKenzie, 844 E.H.C., Johnston, P.R., 2009. A polyphasic approach for studying Colletotrichum. 845 Fungal Diversity 39, 183–124. 846 Cannon, P.F., Damm, U., Johnston, P., Weir, B.S., 2012. Colletotrichum – Current 847 status and future directions. Studies in Mycology 73, 181–213. 848 Cao, X.; Xu, X.; Che, H.; West, J.S.; Luo, D., 2019. Three Colletotrichum Species, 849 Including a New Species, are Associated to Leaf Anthracnose of Rubber Tree in 850 Hainan, China. Plant Disease 103, 117–124. 851 Corda, A.C.I., 1831. Die Pilze Deutschlands. In: Deutschlands Flora in Abbildungen 852 nach der Natur mit Beschreibungen (Sturm, J, ed.). Sturm, Nürnberg vol. 3, Abt. 853 12: 33–64, tab. 21–32. 854 Costa, J.F.O., Kamei, S.H., Silva, J.R.A., Miranada, A.R.G.S., Netto, M.B., Silva, 855 S.J.C., Correia, K.C., Lima, G.S.A., Assunção, I.P., 2018. Species diversity of 856 Colletotrichum infecting Annona spp. in Brazil. European Journal of Plant 857 Pathology. https://doi.org/10.1007/s10658-018-01630-w. 858 Crouch, J.A., Clarke, B.B., Hillman, B.I, 2009a. What is the value of ITS sequence 859 data in Colletotrichum systematics and species diagnosis? A case study using 860 the falcate-spored graminicolous Colletotrichum group. Mycologia 101, 648–656. 861 Crouch, J.A., Beirn, L.A., Cortese, L.M., Bonos, S.A., Clarke, B.B., 2009b. 862 Anthracnose disease of switchgrass caused by the novel fungal species 863 Colletotrichum navitas. Mycological Research 113, 1411–1421 864 Crouch, J.A., Tredway, L.P., Clarke, B.B. and Hillman, B.I., 2009c. Phylogenetic and 865 population genetic divergence correspond with habitat for the pathogen bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 866 Colletotrichum cereale and allied taxa across diverse grass communities. 867 Molecular Ecology 18, 123–135. 868 Crouch, J.A., Tomaso-Peterson, M., 2012. Anthracnose disease of centipedegrass 869 turf caused by Colletotrichum eremochloae, a new fungal species closely related 870 to Colletotrichum sublineola. Mycologia 104, 1085–1096. 871 Cummings, M.P., Neel, M.C., Shaw, K.L., 2008. A genealogical approach to 872 quantifying lineage divergence. Evolution 62, 2411–2422. 873 Damm, U., Woudenberg, J.H.C., Cannon, P.F., Crous, P.W., 2009. Colletotrichum 874 species with curved conidia from herbaceous hosts. Fungal Diversity 39, 45–87. 875 Damm, U., Cannon, P.F., Liu, F., Barreto, R.W., Guatimosim, E, Crous,P.W., 2013. 876 The Colletotrichum orbiculare species complex: Important pathogens of field 877 crops and weeds. Fungal Diversity 61, 29–59. 878 Damm, U., Cannon, P.F., Woudenberg, J.H.C., Crous, P., 2012a. The Colletotrichum 879 acutatum species complex. Studies in Mycology 73, 37–113. 880 Damm, U., Cannon, P.F., Woudenberg, J.H.C., Johnston, P.R., Weir, B.S., Tan, Y.P., 881 Shivas, R.G., Crous, P.W., 2012b. The Colletotrichum boninense species 882 complex. Studies in Mycology 73:1–36. 883 Damm, U., Sato, T., Alizadeh, A., Groenewald, J.Z., Crous,P.W., 2019. The 884 Colletotrichum dracaenophilum, C. magnum and C. orchidearum species 885 complexes. Studies in Mycology 92, 1–46. 886 Dean, R., Van Kan, J.A.L., Pretorius, Z.A., Hammond-Kosack, K.E., Di Pietro, A., 887 Spanu, P.D., Rudd, J.J., Dickman, M., Kahmann, R., Ellis, J., Foster, G.D., 2012. 888 The Top 10 fungal pathogens in molecular plant pathology. Molecular Plant 889 Pathology 13, 414–430. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 890 Dettman, J.R., Jacobson, D.J., Taylor, J.W., 2003. A multilocus genealogical 891 approach to phylogenetic species recognition in the model eukaryote 892 Neurospora. Evolution 57, 2703–2720. 893 Diao, Y.-Z., Zhang, C., Liu, F., Wang, W.-Z., Liu, L., Cai, L., Liu, X.-L., 2017. 894 Colletotrichum species causing anthracnose disease of chili in China. Persoonia 895 38, 20–37. 896 Doyle, V.P., Oudemans, P.V., Rehner, S.A., Litt, A., 2013. Habitat and host indicate 897 lineage identity in Colletotrichum gloeosporioides s. l. from wild and agricultural 898 landscapes in North America. PLoS ONE 8, e62394. 899 Du, M., Schardl, C., Nuckles, E., Vaillancourt, L., 2005. Using mating-type gene 900 sequences for improved phylogenetic resolution of Collectotrichum species 901 complexes. Mycologia 97, 641–658. 902 Fong, J.J., Fujita, M.K., 2011. Evaluating phylogenetic informativeness and data-type 903 usage for new protein-coding genes across Vertebrata. Molecular Phylogenetics 904 and Evolution 61, 300–307. 905 Fu, M., Crous, P.W., Bai, Q., Zhang, P.F., Xiang, J., Guo, Y.S., Zhao, F.F., Yang, 906 M.M., Hong, N., Xu, W.X. and Wang, G.P., 2019. Colletotrichum species 907 associated with anthracnose of Pyrus spp. in China. Persoonia 42, 1–35. 908 Hebert, P.D., Cywinska, A., Ball, S.L., 2003. Biological identifications through DNA 909 barcodes. Proceedings of the Royal Society of London Series B 270, 313–321. 910 Hyde KD, Cai L, McKenzie EHC, Yang YL, Zhang JZ, Prihastuti, H., 2009. 911 Colletotrichum: a catalogue of confusion. Fungal Diversity 39, 1–17. 912 Hyde, K.D., Udayanga, D., Manamgoda, D.S., Tedersoo, L., Larsson, E., Abarenkov 913 K., Bertrand, Y.J.K., Oxelman, B., Hartmann, M., Kauserud, H., Ryberg, M., 914 Kristiansson, E., Nilsson, R.H., 2013. Incorporating molecular data in fungal bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 915 systematics: A guide for aspiring researchers. Current Research in 916 Environmental and Applied Mycology 3, 1–32. 917 Jayawardena, R.S., Yan, J., Hyde, K.D., Zhang, G.,2016. Morphological and 918 molecular characterization of Colletotrichum species of strawberry in China. 919 Mycosphere 7, 1147–1163. 920 Katoh, K., Misawa, K., Kuma, K., Miyata, T., 2002. MAFFT: a novel method for rapid 921 multiple sequence alignment based on fast Fourier transform. Nucleic Acids 922 Research 30, 3059–3066. 923 Katoh, K., Standley, D.M, 2013. MAFFT multiple sequence alignment software 924 version 7: Improvements in performance and usability. Molecular Biology and 925 Evolution 30, 772–780. 926 Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: Molecular Evolutionary Genetics 927 Analysis version 7.0 for bigger datasets. Molecular Biology and Evolution 33, 928 1870–1874. 929 Larget, B., Kotha, S.K., Dewey, C.N., Ané, C., 2010. BUCKy: Gene tree/species tree 930 reconciliation with the Bayesian concordance analysis. Bioinformatics 26, 2910– 931 2911. 932 Librado, P., Rozas, J.,2009. DnaSP v5: a software for comprehensive analysis of 933 DNA polymorphism data. Bioinformatics 25, 1451–1452. 934 Lima, N.B., Batista, M.V.A., Morais Jr, M.A., Barbosa, M.A.G., Michereff, S.J., Hyde, 935 K.D., Câmara, M.P.S., 2013. Five Colletotrichum species are responsible for 936 mango anthracnose in northeastern Brazil. Fungal Diversity 61:75–88. 937 Liu, F., Cai, L., Crous, P.W., Damm, U., 2014. The Colletotrichum gigasporum 938 species complex. Persoonia 33, 83–97. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 939 Liu, F., Wang, M., Damm, U., Crous, P.W., Cai, L., 2016. Species boundaries in plant 940 pathogenic fungi: A Colletotrichum case study. BMC Evolutionary Biology 16, 81. 941 Liu, F., Weir, B.S., Damm, U., Crous, P.W., Wang, Y., Liu, B., Wang, M., Zhang, M., 942 Cai, L., 2015. Unravelling Colletotrichum species associated with Camellia: 943 Employing ApMat and GS loci to resolve species in the C. gloeosporioides 944 complex. Persoonia 35, 63–86. 945 Lopez-Giraldez, F., Townsend, J.P., 2011. PhyDesign: An online application for 946 profiling phylogenetic informativeness. BMC Evolutionary Biology 11, 152. 947 Magain, N., Miadlikowska, J., Mueller, O., Gajdeczka, M., Truong, C., Salamov, A.A., 948 Dubchak, I., Grigoriev, I.V., Goffinet, B., Sérusiaux, E., Lutzoni, F., 2017. 949 Conserved genomic collinearity as a source of broadly applicable, fast evolving, 950 markers to resolve species complexes: A case study using the lichen-forming 951 genus Peltigera section Polydactylon. Molecular Phylogenetics and Evolution 952 117, 10–29. 953 Marin-Felix, Y., Groenewald, J.Z., Cai, L., Chen, Q., Marincowitz, S., Barnes, I., 954 Bensch, K., Braun, U., Camporesi, E., Damm, U., de Beer, Z.W., Dissanayake, 955 A., Edwards, J., Giraldo, A., Hernandez-Restrepo, M., Hyde, K.D., Jayawardena, 956 R.S., Lombard, L., Luangsa-Ard, J., McTaggart, A.R., Rossman, A.Y., Sandoval- 957 Denis, M., Shen, M., Shivas, R.G., Tan, Y.P., van der Linde, E.J., Wingfield, M.J., 958 Wood, A.R., Zhang, J.Q., Zhang, Y., Crous, P.W., 2017. Genera of 959 phytopathogenic fungi: GOPHY 1. Studies in Mycology 86, 99–216. 960 Mills, P.R., Hodson, A., Brown, A.E., 1992. Molecular differentiation of Colletotrichum 961 gloeosporioides isolates infecting tropical fruits, in: Bailey, J.A., Jeger, M.J. 962 (Eds.), Colletotrichum: Biology, Pathology and Control, CABI, Wallingford, 269– 963 288. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 964 Moriwaki,J., Tsukiboshi,T., 2009. Colletotrichum echinochloae, a new species on 965 Japanese barnyard millet (Echinochloa utilis). Mycoscience 50, 273–280. 966 Niu, X., Gao, H., Qi, J., Chen, M., Tao, A., Xu, J., Dai, Z., Su, J., 2016. Colletotrichum 967 species associated with jute (Corchorus capsularis L.) anthracnose in 968 southeastern China. Scientific Reports 6, 25179. 969 O'Connell, R.J., Thon, M.R., Hacquard, S., Amyotte, S.G., Kleemann, J., Torres, 970 M.F., Damm, U., Buiate, E.A., Epstein, L., Alkan, N., Altmuller, J., Alvarado- 971 Balderrama, L., Bauser, C.A., Becker, C., Birren, B.W., Chen, Z., Choi, J., 972 Crouch, J.A., Duvick, J.P., Farman, M.A., Gan, P., Heiman, D., Henrissat, B., 973 Howard, R.J., Kabbage, M., Koch, C., Kracher, B., Kubo, Y., Law, A.D., Lebrun, 974 M.H., Lee, Y.H., Miyara, I., Moore, N., Neumann, U., Nordstrom, K., Panaccione, 975 D.G., Panstruga, R., Place, M., Proctor, R.H., Prusky, D., Rech, G., Reinhardt, 976 R., Rollins, J.A., Rounsley, S., Schardl, C.L., Schwartz, D.C., Shenoy, N., 977 Shirasu, K., Sikhakolli, U.R., Stuber, K., Sukno, S.A., Sweigard, J.A., Takano, Y., 978 Takahara, H., Trail, F., Van Der Does, H.C., Voll, L.M., Will, I., Young, S., Zeng, 979 Q., Zhang, J., Zhou, S., Dickman, M.B., Schulze-Lefert, P., Van Themaat, E.V., 980 Ma, L.J., Vaillancourt,L.J., 2012. Lifestyle transitions in plant pathogenic 981 Colletotrichum fungi deciphered by genome and transcriptome analyses. Nature 982 Genetics 44, 1060–1065. 983 Oliveira, L.F.M., Feijó, F.M., Mendes, A.L.S.F., Neto, J.D.V., Netto, M.S.B., 984 Assunção, I.P., Lima, G.S.A., 2018. Identification of Colletotrichum species 985 associated with brown spot of cactus prickly pear in Brazil. Tropicala Plant 986 Pathology 43, 247–253. 987 Paradis, E., Claude, J., Strimmer, K., 2004. APE: Analyses of phylogenetics and 988 evolution in R language. Bioinformatics 20, 289–290. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 989 Pond, S.L., Frost, S.D., Muse, S.V., 2005. HyPhy: hypothesis testing using 990 phylogenies. Bioinformatics 21, 676–679. 991 R Core Team. 2017. R: A language and environment for statistical computing. R 992 Foundation for Statistical Computing, Vienna, Austria. 993 Rojas, E.I., Rehner, S.A., Samuels, G.J., Van Bael, S.A., Herre, E.A., Cannon, P., 994 Chen, R., Pang, J., Wang, R., Zhang, Y., Peng, Y.Q., Sha, T., 2010. 995 Colletotrichum gloeosporioides s.l. associated with Theobroma cacao and other 996 plants in Panamá: Multilocus phylogenies distinguish host-associated pathogens 997 from asymptomatic endophytes. Mycologia 102,1318–1338. 998 Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., 999 Larget, B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes v. 3.2: 1000 Efficient Bayesian phylogenetic inference and model choice across a large model 1001 space. Systematic Biology 61, 539–542. 1002 Sakalidis, M.L., Hardy, G.E.S.J., Burgess, T.I., 2011. Use of the Genealogical Sorting 1003 Index (GSI) to delineate species boundaries in the Neofusicoccum parvum- 1004 Neofusicoccum ribis species complex. Molecular Phylogenetics and Evolution, 1005 60, 333–344. 1006 Samarakoon, M.C., Peršoh, D., Hyde, K.D., Bulgakov, T.S., Manawasinghe, I.S., 1007 Jayawardena, R.S., Promputtha, I., 2018. Colletotrichum acidae sp. nov. from 1008 northern Thailand and a new record of C. dematium on Iris sp. Mycosphere 9: 1009 583–597. 1010 Schmitt,.I, Crespo, A., Divakar, P.K., Fankhauser, J.D., Herman-Sackett, E., Kalb, K., 1011 Nelsen, M.P., Nelson, N.A., Rivas-Plata, E., Shimp, A.D.,Widhelm, T., Lumbsch, 1012 H.T., 2009. New primers for promising single-copy genes in fungal phylogenetics 1013 and systematics. Persoonia 23, 35–40. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1014 Sela, I., Ashkenazy, H., Katoh, K., Pupko, T., 2015. GUIDANCE2: Accurate detection 1015 of unreliable alignment regions accounting for the uncertainty of multiple 1016 parameters. Nucleic Acids Research 43, W7–W14. 1017 Sharma, G., Kumar, N., Weir, B.S., Hyde, K.D., Shenoy, B.D., 2013. The ApMat 1018 marker can resolve Colletotrichum species: A case study with Mangifera indica. 1019 Fungal Diversity 61,117–38. 1020 Sharma, G., Pinnaka, A.K., Belle, D.S., 2015. Resolving the Colletotrichum siamense 1021 species complex using ApMat marker. Fungal Diversity 71, 247–64. 1022 Sharma, G., Maymon, M., Freeman, S., 2017. Epidemiology, pathology and 1023 identification of Colletotrichum including a novel species associated with avocado 1024 (Persea americana) anthracnose in Israel. Scientific Reports 7, 15839. 1025 Silva, D.N., Talhinas, P., Várzea, V., Cai, L., Paulo, O.S., Batista, D., 2012. 1026 Application of the Apn2/MAT locus to improve the systematics of the 1027 Colletotrichum gloeosporioides complex: An example from coffee (Coffea spp.) 1028 hosts. Mycologia 104, 396–409. 1029 Silva, J.R.A, Chaves, T.P., Silva,a A.R.G., Barbosa, L.F., Costa, F.F.O., Ramos- 1030 Sobrinho, R., Teixeira, R.R.O., Silva, S.J.C., Lima, G.S.A., Assunção, I.P., 2017. 1031 Molecular and morpho-cultural characterization of Colletotrichum spp. associated 1032 with anthracnose on Capsicum spp. in northeastern Brazil. Tropical Plant 1033 Pathology 42, 315–319. 1034 Sousa, E.S., Silva, J.R.A., Assunção, I.P., Melo, M.P., Feijó, F.M., Matos, K.S., Lima, 1035 G.S.A., Beserra Jr, J.E.A., 2018. Colletotrichum species causing anthracnose on 1036 lima bean in Brazil. Tropical Plant Pathology 43, 78–84. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1037 Sreenivasaprasad, S., Brown, A.E., Mills, P.R., 1992. DNA sequence variation and 1038 interrelationship among Colletotrichum species causing strawberry anthracnose. 1039 Physiological and Molecular Plant Pathology 41, 265–281. 1040 Stawatakis, A., 2014. RAxML version 8: A tool for Phylogenetic Analysis and Post- 1041 Analysis of Large Phylogenies. Bioinformatics 30, 1312–1313. 1042 Sutton, B.C., 1980. The Coelomycetes. Fungi Imperfecti with Pycnidia, Acervuli and 1043 Stromata. CABI, Kew 1044 Tao, G., Liu, Z.Y., Liu, F., Gao, Y.H., Cai,L., 2013. Endophytic Colletotrichum species 1045 from Bletilla ochracea (Orchidaceae), with descriptions of seven new speices. 1046 Fungal Diversity 61, 139–164. 1047 Taylor, J.W., Jacobson, D.J., Kroken, S., Kasuga, T., Geiser, D.M., Hibbett, D.S., 1048 Fisher, M.C., 2000. Phylogenetic species recognition and species concepts in 1049 fungi. Fungal Genetics and Biology 31, 21–32. 1050 Vaidya, G., Lohman, D. J., Meier, R., 2011. SequenceMatrix: concatenation software 1051 for the fast assembly of multi-gene datasets with character set and codon 1052 information. Cladistics 27, 171–180. 1053 Veloso, J.S., Câmara, M.P.S., Lima, W.G., Michereff, S.J., Doyle, V.P.. (2018). Why 1054 species delimitation matters for fungal ecology: Colletotrichum diversity on wild 1055 and cultivated cashew in Brazil. Fungal Biology 122, 677–691. 1056 Vieira, W.A., Michereff, S.J., de Morais Jr, M.A., Hyde, K.D., Câmara, M.P.S., 2014. 1057 Endophytic species of Colletotrichum associated with mango in northeastern 1058 Brazil. Fungal Diversity 67, 181–202. 1059 Vieira, W.A.S., Lima, W.G., Nascimento, E.S., Michereff, S.J., Câmara, M.P.S., 1060 Doyle, V.P., 2017. The impact of phenotypic and molecular data on the inference 1061 of Colletotrichum diversity associated with Musa. Mycologia 109, 912–934. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1062 von Arx, J.A., 1957. Die Arten der Gattung Colletotrichum Cda. Phytopathologische 1063 Zeitschrift 29, 413–468. 1064 Wang, Q.T., Liu, X.T., Ma, H.Y., Shen, X.Y., Hou, C.L., 2019. Colletotrichum 1065 yulongense sp. nov. and C. rhombiforme isolated as endophytes from Vaccinium 1066 dunalianum var. urophyllum in China. Phytotaxa 394, 285–298. 1067 Weir, B.S., Johnston, P.R., Damm, U., 2012. The Colletotrichum gloeosporioides 1068 species complex. Studies in Mycology 73,115–180. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. Table 1. Alignment and phylogenetic informativeness profile statistics for markers used in Colletotrichum by species complex. Table 1. Parsymony 1 1 1 1 -6 2 3 C. acutatum s. l. Length Invariable Variable Singletons PIV (10 ) Pimax informative ACT 239 163 76 24 52 15.36 0.99 CHS-1 251 216 35 5 30 8.56 0.78 GAPDH 272 165 94 18 76 19.95 0.78 HIS3 386 298 88 18 70 24.96 0.67 nrITS 454 427 26 7 19 5.41 0.99 TUB2 481 386 95 24 71 19.7 0.99 C. boninense s. l. ACT 277 199 78 15 63 29.07 0.49 CAL 438 293 142 16 126 52.86 0.58 CHS-1 277 228 49 4 45 22.39 0.48 GAPDH 247 163 83 9 74 47.6 0.24 HIS3 389 292 93 11 82 46.96 0.32 nrITS 537 490 47 11 36 18.41 0.39 TUB2 483 345 137 23 114 48.35 0.63 C. dematium s. l. ACT 235 188 46 1 45 44.65 0.09 CHS-1 251 215 36 6 30 21.48 0.5 GAPDH 264 140 122 13 109 237.56 0.05 HIS3 371 301 70 7 63 183.1 0.05 nrITS 517 492 25 1 24 32.94 0.06 TUB2 497 389 105 10 95 118.42 0.06 C. destructivum s. l. ACT 263 218 44 20 24 18 0.66 CHS-1 280 257 23 11 12 10.13 0.63 GAPDH 225 149 47 16 31 27.35 0.28 HIS3 389 318 71 18 53 38.75 0.37 nrITS 557 526 26 13 13 12.41 0.17 TUB2 514 432 77 33 44 32.49 0.73 C. dracaenophilum s. l. ACT 254 205 49 6 43 31.72 0.44 CHS-1 282 257 25 5 20 18.53 0.46 GAPDH 273 192 81 14 67 52.08 0.41 HIS3 417 354 62 7 55 35.74 0.45 nrITS 541 504 37 13 24 28.58 0.39 TUB2 492 411 81 9 72 58.22 0.4 C. gigasporum s. l. ACT 280 239 37 17 20 23.82 0.19 CAL 675 543 131 56 75 63.14 0.44 CHS-1 299 266 33 15 18 26.38 0.11 GAPDH 290 193 94 51 43 61.02 0.28 GS 702 521 176 92 84 105.17 0.10 HIS3 416 362 53 22 31 45.36 0.12 nrITS 545 500 39 15 24 34.85 0.11 TUB2 537 451 86 40 46 58.50 0.15 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. gloeosporioides s. l. APN2 735 624 108 18 90 73.91 0.17 APN2/MAT-IGS 579 378 201 4 197 149.81 0.19 CAL 648 578 70 1 69 48.76 0.27 GAP2-IGS 708 569 135 14 121 79.71 0.42 GAPDH 767 734 33 2 31 24.51 0.33 GS 665 571 87 24 63 43.83 0.39 TUB2 1231 1116 115 10 105 67.23 0.49 C. graminicola s. l. ACT 269 174 82 52 30 30.25 0.68 CHS-1 280 237 43 24 19 17.06 0.55 nrITS 454 375 70 37 33 30.38 0.23 TUB2 500 350 145 89 56 58.96 0.46 C. magnum s. l. ACT 276 260 16 13 3 11.16 0.40 CHS-1 257 247 10 7 3 11.16 0.40 GAPDH 241 215 26 14 12 13.35 0.42 HIS3 399 360 39 28 11 26 0.44 nrITS 539 533 6 6 0 4.16 0.40 TUB2 536 500 36 24 12 23.15 0.52 C. orbiculare s. l. ACT 226 202 21 5 16 18.37 0.36 CHS-1 280 268 12 2 10 6.16 0.65 GAPDH 243 203 40 8 32 32.01 0.17 GS 954 810 134 41 93 69.13 0.53 HIS3 386 334 50 10 40 36.51 0.13 nrITS 529 514 15 4 11 6.38 0.52 TUB2 467 431 36 9 27 19.75 0.55 C. orchidearum s. l. ACT 276 244 29 17 12 7.06 0.52 CHS-1 265 243 22 6 16 11.46 0.47 GAPDH 242 195 47 37 10 7.99 0.77 HIS3 402 343 59 31 28 16.48 0.67 nrITS 539 521 18 10 8 5.81 0.57 TUB2 540 460 79 51 28 17.44 0.80 C. spaethianum s. l. ACT 210 152 56 28 28 19.44 0.77 GAPDH 240 132 105 28 77 32.53 0.47 nrITS 482 470 10 4 6 8.66 0.73 TUB2 443 351 92 44 48 33.26 0.62 C. truncatum s. l. ACT 215 179 36 6 30 22.89 0.41 GAPDH 276 186 90 6 84 56.86 0.46 nrITS 479 459 20 4 16 10.17 0.53 TUB2 434 353 76 9 67 33.14 0.53 1 - number of base pairs. 2 - PIV, phylogenetic informativeness values on a per net basis 3 - PImax, optimal divergence time. Values are expressed in arbitrary units. 70 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1072 FIGURE CAPTIONS 1073 Fig. 1. Barcode gap and distance overlap between the intra- and inter-specific 1074 distances. Values were calculated based on the intra- and inter-specific distances 1075 frequencies distribution of each Colletotrichum species complex. 1077 Fig. 2. Ultrametric trees and net phylogenetic informativeness profiles of markers 1078 used for phylogenetic studies of 13 Colletotrichum species complexes. Values on the 1079 X-axes correspond to the relative timescale (0—1) based on the root-to-tip distance. -6 1080 Values on the Y-axes represent net phylogenetic informativeness values (10 ) in 1081 arbitrary units. 1083 Fig. 3. Heat map of the Genealogical Sorting Indices (GSI) by Colletotrichum species 1084 complex. GSIs of 1000 bootstrap trees were calculated with 100 permutations. Rows 1085 correspond to species, and columns correspond to individual markers and 1086 concatenated datasets (all markers and best markers). Asterisks represent the best 1087 markers for each complex. 1089 Fig. 4. Primary concordance trees resulting from the Bayesian concordance analyses 1090 including isolates from the C. acutatum complex. A. All markers (ACT, CHS-1, 1091 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1092 Concordance factors are shown above the branches that were resolved by at least 1093 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1095 Fig. 5. Primary concordance trees resulting from the Bayesian concordance analyses 1096 including isolates from C. boninense complex. A. All markers (ACT, CAL, CHS-1, 1097 GAPDH, HIS3, ITS and TUB2). B. Best markers (CAL, GAPDH and TUB2). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1098 Concordance factors are shown above the branches that were resolved by at least 1099 one marker (≥0.14 for all markers and ≥0.33 for the best markers). 1101 Fig. 6. Primary concordance trees resulting from the Bayesian concordance analyses 1102 including isolates from the C. dematium complex. A. All markers (ACT, CHS-1, 1103 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1104 Concordance factors are shown above the branches that were resolved by at least 1105 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1107 Fig. 7. Primary concordance trees resulting from the Bayesian concordance analyses 1108 including isolates from the C. destructivum complex. A. All markers (ACT, CHS-1, 1109 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1110 Concordance factors are shown above the branches that were resolved by at least 1111 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1113 Fig. 8. Primary concordance trees resulting from the Bayesian concordance analyses 1114 including isolates from the C. dracaenophilum complex. A. All markers (ACT, CHS-1, 1115 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1116 Concordance factors are shown above the branches that were resolved by at least 1117 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1119 Fig. 9. Primary concordance trees resulting from the Bayesian concordance analyses 1120 including isolates from the C. gigasporum complex. A. All markers (ACT, CAL, CHS- 1121 1, GAPDH, GS, HIS3, ITS and TUB2). B. Best markers (CAL, GAPDH and GS). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1122 Concordance factors are shown above the branches that were resolved by at least 1123 one marker (≥0.13 for all markers and ≥0.33 for the best markers). 1125 Fig. 10. Primary concordance trees resulting from the Bayesian concordance 1126 analyses including isolates from the C. gloeosporioides complex. A. All markers 1127 (APN2, APN2/MAT-IGS, CAL, GAPDH, GAP2-IGS, GS and TUB2). B. Best markers 1128 (APN2, APN2/MAT-IGS and GAP2-IGS). Concordance factors are shown above the 1129 branches that were resolved by at least one marker (≥0.14 for all markers and ≥0.33 1130 for the best markers). 1132 Fig. 11. Primary concordance trees resulting from the Bayesian concordance 1133 analyses including isolates from the C. magnum complex. A. All markers (ACT, CHS- 1134 1, GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1135 Concordance factors are shown above the branches that were resolved by at least 1136 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1138 Fig. 12. Primary concordance trees resulting from the Bayesian concordance 1139 analyses including isolates from the C. orbiculare complex. A. All markers (ACT, 1140 CHS-1, GAPDH, GS, HIS3, ITS and TUB2). B. Best markers (HIS3, GAPDH and 1141 GS). Concordance factors are shown above the branches that were resolved by at 1142 least one marker (≥0.14 for all markers and ≥0.33 for the best markers). 1144 Fig. 13. Primary concordance trees resulting from the Bayesian concordance 1145 analyses including isolates from the C. orchidearum complex. A. All markers (ACT, 1146 CHS-1, GAPDH, HIS3, ITS and TUB2). B. Best markers (CHS-1, HIS3 and TUB2). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1147 Concordance factors are shown above the branches that were resolved by at least 1148 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1150 Fig. 14. Primary concordance trees resulting from the Bayesian concordance 1151 analyses including isolates from the C. spaethianum complex. A. All markers (ACT, 1152 GAPDH, ITS and TUB2). B. Best markers (ACT, GAPDH, and TUB2). Concordance 1153 factors are shown above the branches that were resolved by at least one marker 1154 (≥0.25 for all markers and ≥0.33 for the best markers). 1156 Fig. 15. Primary concordance trees resulting from the Bayesian concordance 1157 analyses including isolates from the C. truncatum complex. A. All markers (ACT, 1158 GAPDH, ITS and TUB2). B. Best markers (ACT, GAPDH, and TUB2). Concordance 1159 factors are shown above the branches that were resolved by at least one marker 1160 (≥0.25 for all markers and ≥0.33 for the best markers). 1163 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. acutatum s. l. C. boninense s. l. C. dematium s. l. C. destructivum s. l. 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 GAPDH GAPDH GAPDH GAPDH CAL ACT ACT HIS3 CHS-1 HIS3 TUB ACT HIS3 TUB2 CHS-1 HIS3 ACT CHS-1 TUB2 CHS-1 TUB2 nrITS nrITS nrITS nrITS 0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.02 0.04 0.06 0.08 0 0,02 0,04 0,06 0,08 0,1 0,12 0 0.05 0.1 0.15 0.2 0.25 0.3 C. dracaenophilum s. l. C. gigasporum s. l. C. gloeosporioides s. l. C. magnum s. l. 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 GS APN2/MAT-IGS GAPDH GAPDH CAL GAP2-IGS ACT HIS3 GAPDH APN2 TUB2 TUB2 TUB2 CAL ACT CHS-1 HIS3 TUB2 CHS-1 ACT CHS-3 GS HIS3 nrITS nrITS GAPDH nrITS 0 0.01 0.02 0.03 0.04 0.05 0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2 0.25 0 0.05 0.1 0.15 C. orchidearum s. l. C. spaethianum s. l. C. truncatum s. l. C. orbiculare s. l. 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 GS HIS3 GAPDH GAPDH HIS3 CHS-1 GAPDH ACT ACT TUB2 ACT GAPDH TUB2 TUB2 TUB2 ACT CHS-1 nrITS nrITS nrITS nrITS 0 0.005 0.01 0.015 0.02 0.025 0.03 0 0.02 0.04 0.06 0.08 0.1 0 0.05 0.1 0.15 0.2 0.25 0 0.05 0.1 0.15 0.2 0.25 Intra/inter -specific Barcode gap distance overlap bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. simmondsii C. paxtonii C. sloanei C. indonesiense C. lacticiphilum C. brisbanense C. cairnsense C. guajavae C. boninense C. scovillei C. nymphaeae C. torulosum C. walleri C. cybidiicola C. cosmi C. oncidii C. chrysantemi C. beeveri C. melonis C. paranaense C. colombiense C. anthirrhinicola C. limetticola C. brassicicola C. costaricense C. fuscum C. paspali C. annellatum C. tamarilloi C. bryoniicola C. lupini C. phyllanthi C. nicholsonii C. vignae C. cuscutae C. karstii C. graminicola C. abscissum C. higginsianum C. petchii C. navitas C. acutatum C. utrechtense C. fioriniae C. limonicola C. susblineola C. tabacum C. rhobiforme C. novae-zelandiae C. eremochloae C. acerbum C. ocimi C. catinaense C. phormi C. cereale C. destructivum C. parsoniae C. kinghornii C. mischanti C. australe C. americae-borealis C. brasiliense C. salicis C. eleusines C. lini C. hippeastri C. pyricola C. jacksonii C. lentis C. constrictum C. johnstonii C. falcatum C. godetiae C. pisicola C. dacrycarpi 36 55 20 49 18 38 15 33 24 37 13 27 20 30 10 22 16 24 7 16 12 18 5 11 8 12 5 4 0 0 0 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 C. acutatum s. l. C. graminicola s. l. C. boninense s. l. C. destructivum s. l. C. tofieldiae C. verruculosum C. orbiculare C. cliviicola C. liriopes C. sidae C. plurivorum C. chrysophilum C. bletilum C. spinosum C. vittalense C. fructicola C. riograndense C. tebeestii C. orchidearum C. musae C. incanum C. trifolii C. musicola C. siamense C. spaethianum C. malvarum C. cattleyicola C. queenslandicum C. lilii C. bidentis C. soojae C. tropicale C. guizhouensis C. lindemuthianum C. theobromicola C. piperis 45 16 40 14 125 35 12 110 10 94 25 9 78 20 7 62 15 5 47 10 3 31 5 15 0 0 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 C. orchidearum s. l. C. gloeosporioides s. l. C. spaethianum s. l. C. orbiculare s. l. C. lineola C. brevisporum C. arxii C. corchorum-capsularis C. dematium C. lobatum C. vietnamense C. fructi C. truncatum C. merremiae C. radicis C. anthrisci C. panamense C. curcumae C. pseudomajus C. circinans C. magnum C. acidae C. spinaceae C. cacao C. gigasporum C. fusisforme 27 110 24 53 199 47 19 77 16 35 124 29 13 55 10 44 8 33 17 5 22 11 2 11 5 0 0 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 C. dematium s. l. C. magnum s. l. C. gigasporum s. l. C. truncatum s. l. C. yunnanense C. dracaenophilum C. coelogynes C. tropicicola GAP2-IGS nrITS ACT CAL APN2 CHS-1 GS TUB2 GAPDH HIS3 5 APN2/MAT-IGS 0.8 0.6 0.4 0.2 0 C. dracaenophilum s. l. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. truncatum s. l. C. dracaenophilum s. l. C. magnum s. l. C. gigasporum s. l. C. acidae C. coelogynes C. gigasporum C. brevisporum C. corchorum− C. dracaenophilum C. vietnamense capsularis C. magnum C. curcumae C. tropicicola C. truncatum C. orchidearum s. l. C. orbiculare s. l. C. dematium s. l. C. spaethianum s. l. C. cliviicola C. lindemuthianum C. anthrisci C. guizhoensis C. malvarum C. musicola C. circinans C. lilii C. orbiculare C. orchidearum C. dematium C. liriopes C. plurivorum C. sidae C. lineola C. spaethianum C. spinosum C. sojae C. spinaceae C. tofieldiae C. trifolii C. vittalense C. boninense s. l. C. destructivum s. l. C. gloeosporioides s. l. C. acutatum s. l. C. americae- C. abscissum C. chrysophilum C. boninense borealis C. acutatum C. brasiliense C. destructivum C. fragariae C. australe C. fuscum C. fructicola C. catinaense C. cairnsense C. colombiense C. higginsianum C. musae C. chrysanthemi C. lentis C. constrictum C. queenslandicum C. siamense C. costaricense C. cymbidiicola C. lini C. tabacum C. tropicale C. fioriniae C. hippeastri C. utrechtense C. karstii C. laticiphilum C. vignae C. lupini C. limonicola C. novae−zelandiae C. meleonis C. oncidii C. nymphaeae C. petchii C. paranaense C. torulosum C. paxtonii C. phormii C. scovillei C. simmondsii C. tamarilloi 0 0.25 0.50 0.75 1 GSI values All markers All markers All markers Best markers Best markers Best markers ACT *ACT ACT CHS-1 *CHS-1 *GAPDH *GAPDH nrITS GAPDH *HIS3 *TUB2 *HIS3 nrITS nrITS *TUB2 *TUB2 All markers All markers All markers Best markers Best markers Best markers ACT ACT ACT CHS-1 CHS.1 *CAL *GAPDH CHS-1 *GAPDH *HIS3 *GAPDH *GS nrITS *HIS3 *HIS3 *TUB2 nrITS nrITS TUB2 TUB2 All markers All markers All markers Best markers Best markers Best markers ACT ACT ACT *CAL CHS-1 CHS-1 CHS-1 *GAPDH *GAPDH *GAPDH *HIS3 *GS *HIS3 nrITS HIS3 *TUB2 nrITS nrITS *TUB2 TUB2 All markers All markers All markers Best markers Best markers Best markers ACT *APN2 *ACT *APN2/MAT-IGS CHS-1 *GAPDH CAL *GAPDH GAPDH nrITS *HIS3 *GAP2-IGS *TUB2 nrITS GS *TUB2 TUB2 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.48 0.73 C. scovillei C. scovillei A 0.37 0.4 B C. guajavae C. guajavae 0.17 0.17 0.58 C. cairnsense C. chrysanthemi 0.21 C. brisbanense 1 0.39 C. cairnsense 0.19 0.33 C. laticiphilum C. cosmi 0.55 0.36 C. indonesiense C. walleri 0.99 0.44 C. cosmi C. nymphaeae Clade 2 0.18 C. walleri 0.74 0.66 0.73 0.21 C. nymphaeae C. paxtonii 0.68 0.73 0.42 0.34 C. simmondsii C. paxtonii Clade 2 0.38 0.37 0.41 C. sloanei 0.17 C. simmondsii 0.33 0.72 C. laticiphilum 0.37 C. sloanei C. indonesiense 0.84 C. chrysanthemi C. brisbanense 0.31 0.73 0.33 C. paranaense C. abscissum 0.33 0.63 0.18 C. limeticola 0.37 0.85 C. costaricense 0.37 C. melonis 0.38 0.35 C. tamarilloi C. costaricense 0.78 0.17 C. lupini 0.74 0.33 C. lupini 0.26 0.18 C. paranaense Clade 1 Clade 1 C. tamarilloi 0.36 1 C. limeticola 0.83 0.38 C. cuscutae C. melonis 0.59 C. abscissum C. cuscutae Clade 3 Clade 3 1 1 C. fioriniae C. fioriniae C. acerbum 1 1 C. acerbum 0.23 C. rhombiforme C. rhombiforme 0.39 0.42 0.44 0.28 C. phormii C. phormii 0.5 0.43 0.49 C. kinghornii C. kinghornii 0.84 0.27 1 0.41 C. australe C. australe Clade 5 Clade 5 C. salicis 0.9 0.95 C. salicis C. johnstonii C. johnstonii 0.69 0.98 0.51 0.67 0.18 C. pyricola C. pyricola 0.42 C. godetiae C. godetiae 0.18 Clade 4 Clade 4 C. acutatum C. acutatum bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.45 0.53 A C. cymbidiicola B C. cymbidiicola 0.34 0.46 0.73 C. oncidii C. oncidii 0.85 1 0.68 0.76 0.4 C. boninense C. boninense 0.56 0.78 0.77 0.85 1 C. torulosum C. torulosum 0.79 1 C. colombiense C. colombiense 0.38 0.46 0.37 0.79 0.77 C. beeveri C. beeveri C. brassicicola 0.21 C. brassicicola 0.88 0.78 C. brasiliense C. constrictum 0.45 1 0.99 C. parsoniae C. dacrycarpi 1 1 C. hippeastri C. brasiliense 0.4 0.52 0.91 C. constrictum C. hippeastri C. dacrycarpi C. parsoniae 0.57 C. catinaense 0.19 0.49 C. karstii 0.69 0.99 C. limonicola 0.26 0.78 C. novae-zelandiae C. phyllanti 0.16 C. annelatum 0.2 C. petchii C. karstii 0.72 0.16 0.5 0.49 C. catinaense 0.36 0.74 0.45 C. phyllanti 0.98 C. limonicola 0.81 C. annelatum 0.38 C. petchii C. novae-zelandiae bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.49 A B 0.62 0.5 0.56 C. dematium C. dematium 0.69 0.35 0.49 0.61 C. lineola C. lineola 0.53 0.58 1 1 C. anthrisci C. anthrisci C. fructi 0.73 0.35 0.56 C. circinans 0.54 1 1 C. circinans 0.49 0.58 C. spinaceae 0.39 C. spinaceae 0.46 C. fructi bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B 0.4 0.35 C. fuscum C. higginsianum 0.34 0.61 0.69 C. vignae C. vignae C. bryoniicola 0.18 C. antirrhinicola 0.29 0.67 0.23 C. fuscum 0.36 0.18 0.35 C. higginsianum 0.41 C. antirrhinicola 0.37 C. bryoniicola 0.27 0.71 C. utrechtense 0.67 C. utrechtense 0.4 0.37 0.34 0.49 0.29 0.39 C. tabacum C. tabacum 0.2 0.99 0.98 C. lentis 0.37 0.41 C. lini 0.39 C. pisicola 0.19 0.28 C. lini 0.53 C. americae 0.28 0.42 0.55 0.83 0.77 C. americae C. destructivum 0.29 C. ocimi 0.99 C. destructivum 0.99 C. lentis 0.37 C. ocimi C. pisicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A 0.41 B C. dracaenophilum 0.88 C. yunnanense C. coelogynes 1 1 C. tropicicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B 0.5 0.73 C. gigasporum C. gigasporum 0.63 0.38 C. pseudomajus C. pseudomajus 1 1 C. vietnamense C. vietnamense 0.37 0.51 0.37 C. radicis C. arxii C. radicis C. arxii bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.52 A B 0.67 C. theobromicola 0.38 C. siamense 0.4 0.65 0.4 1 0.33 C. musae 0.27 0.77 C. queenslandicum 0.26 0.3 0.59 C. siamense 0.56 0.3 C.theobromicola 0.71 1 C. tropicale 0.85 C. queenslandicum 0.27 0.99 0.71 C. tropicale C. chrysophilum 0.78 0.21 0.3 C. chrysophilum C. fructicola 0.88 0.9 0.45 C. musae 0.9 C. fructicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B 0.33 0.46 0.23 0.4 C. brevisporum 0.55 0.67 0.43 0.51 C. lobatum C. merremiae 0.28 0.39 0.72 C. magnum 0.47 0.5 C. panamense C. cacao bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B C. sidae 0.33 0.52 C. orbiculare 0.66 0.33 C. tebeestii 0.27 0.79 0.17 C. spinosum 0.24 0.44 C. sidae 0.38 0.36 C. tebeestii 0.39 C. orbiculare 0.44 0.65 C. spinosum 0.59 0.44 0.27 0.34 0.72 1 C. malvarum C. malvarum 0.73 0.88 0.21 0.47 0.39 C. trifolii C. trifolii 0.18 0.17 0.32 0.62 0.62 0.71 C. lindemuthianum C. lindemuthianum 0.32 1 1 0.33 C. bidentis C. bidentis bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. cattleyicola A B C. piperis 0.33 0.67 0.71 1 C. sojae 0.5 C. pseudomajus 0.67 1 0.99 C. musicola 0.5 0.67 0.77 1 C. orchidearum 0.74 1 C. vittalense 0.75 1 0.62 0.73 C. plurivorum 0.74 0.81 0.73 0.73 C. cliviicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.55 0.71 A B 0.54 0.7 C. spaethianum 0.37 0.5 0.99 0.99 C. lilii 0.75 1 0.5 0.66 0.27 0.36 0.56 0.64 0.75 C. guizhouensis 1 0.94 0.54 C. incanum C. riograndense 0.6 0.73 0.32 0.35 C. tofieldiae 0.99 0.98 0.82 0.77 C. verruculosum 0.99 1 0.58 0.72 0.28 0.35 0.62 C. liriopes 0.77 0.46 C. bletilum bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.39 A B 0.33 0.25 C. corchorum- 0.55 capsularis C. truncatum 0.35 0.25 0.33 C. corchorum- 0.4 C. truncatum capsularis 0.96 0.97 0.36 C. truncatum 0.29 C. curcumae C. curcumae 0.83 C. acidae C. acidae 0.98 0.9 C. fusiforme C. fusiforme http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png bioRxiv bioRxiv

Loading next page...
 
/lp/biorxiv/optimal-markers-for-the-identification-of-colletotrichum-species-EYmlFo2cve

References (81)

Publisher
bioRxiv
Copyright
© 2019, Posted by Cold Spring Harbor Laboratory. This pre-print is available under a Creative Commons License (Attribution-NonCommercial-NoDerivs 4.0 International), CC BY-NC-ND 4.0, as described at http://creativecommons.org/licenses/by-nc-nd/4.0/
DOI
10.1101/659177
Publisher site
See Article on Publisher Site

Abstract

bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1 Optimal markers for the identification of Colletotrichum species 3 Willie Anderson dos Santos Vieira 4 Priscila Alves Bezerra 5 Anthony Carlos da Silva 6 Josiene Silva Veloso 7 Marcos Paz Saraiva Câmara 8 Vinson Patrick Doyle 10 Universidade Federal Rural de Pernambuco – UFRPE. Departamento de 11 Agronomia. Rua Manuel de Medeiros, s/n - Dois Irmãos, Recife – Pernambuco, 12 52171-900. 13 Department of Plant Pathology and Crop Physiology, Louisiana State University – 14 LSU, AgCenter, Baton Rouge, Louisiana, United States of America, 70808. 16 E-mail (in the author’s order): andersonvieira12@gmail.com; 17 priscilaalvesbezerra@gmail.com; anthonycarlos17@hotmail.com; 18 josieneveloso@yahoo.com.br; marcos.camara@ufrpe.br; vdoyle@agcenter.lsu.edu. 20 Corresponding author: Willie Anderson dos Santos Vieira 21 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 22 ABSTRACT 24 Colletotrichum is among the most important genera of fungal plant pathogens. 25 Molecular phylogenetic studies over the last decade have resulted in a much better 26 understanding of the evolutionary relationships and species boundaries within the 27 genus. There are now approximately 200 species accepted, most of which are 28 distributed among 13 species complexes. Given their prominence on agricultural 29 crops around the world, rapid identification of a large collection of Colletotrichum 30 isolates is routinely needed by plant pathologists, regulatory officials, and fungal 31 biologists. However, there is no agreement on the best molecular markers to 32 discriminate species in each species complex. Here we calculate the barcode gap 33 distance and intra/inter-specific distance overlap to evaluate each of the most 34 commonly applied molecular markers for their utility as a barcode for species 35 identification. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), histone-3 36 (HIS3), DNA lyase (APN2), intergenic spacer between DNA lyase and the mating- 37 type locus MAT1-2-1 (APN2/MAT-IGS), and intergenic spacer between GAPDH and 38 a hypothetical protein (GAP2-IGS) have the properties of good barcodes, whereas 39 sequences of actin (ACT), chitin synthase (CHS-1) and nuclear rDNA internal 40 transcribed spacers (nrITS) are not able to distinguish most species. Finally, we 41 assessed the utility of these markers for phylogenetic studies using phylogenetic 42 informativeness profiling, the genealogical sorting index (GSI), and Bayesian 43 concordance analyses (BCA). Although GAPDH, HIS3 and β-tubulin (TUB2) were 44 frequently among the best markers, there was not a single set of markers that were 45 best for all species complexes. Eliminating markers with low phylogenetic signal 46 tends to decrease uncertainty in the topology, regardless of species complex, and bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 47 leads to a larger proportion of markers that support each lineage in the Bayesian 48 concordance analyses. Finally, we reconstruct the phylogeny of each species 49 complex using a minimal set of phylogenetic markers with the strongest phylogenetic 50 signal and find the majority of species are strongly supported as monophyletic. 52 KEYWORDS 53 Accuracy 54 Anthracnose 55 Barcoding 56 Phylogenetic informativeness 57 Standardization 60 1. INTRODUCTION 61 Colletotrichum is among the largest groups of phytopathogenic fungi and 62 includes the causal agents of anthracnose and other diseases on seeds, stems, 63 leaves and fruits of important temperate and tropical crops (Cai et al., 2009, Cannon 64 et al., 2012). It is also among the most common genera of endophytic fungi, fungi 65 that live within plant organs without producing any symptoms of disease (Cannon et 66 al., 2012). Due to its economic and scientific importance, Colletotrichum was ranked 67 as the eighth most important phytopathogenic fungus in the world by plant 68 pathologists (Dean et al., 2012). 69 Species identification is necessary to understand disease epidemiology and 70 develop strategies to control the disease successfully (Cai et al., 2009). However, 71 Colletotrichum taxonomy and systematics has been a challenge since the genus was bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 72 introduced by Corda (1831). Colletotrichum species were historically circumscribed 73 on the basis of phenotypic features and a strong emphasis on the host species from 74 which the specimens were isolated under the assumption of host specificity, which 75 led to more than 900 species being recognized until revisionary work more than a 76 century after its introduction (von Arx, 1957; Sutton, 1980). Colletotrichum 77 identification based on morphological characters is problematic due to plasticity and 78 variation induced by experimental conditions (Vieira et al. 2017), and all life stages 79 are not frequently produced in culture (Samarakoon et al., 2018). The absence of 80 stable phenotypic characters has limited our understanding of phylogenetic 81 relationships within Colletotrichum and made the recognition of species boundaries 82 unreliable and confusing (Cai et al., 2009). To address this problem, Cai et al. (2009) 83 proposed a guideline for Colletotrichum species recognition based on a polyphasic 84 approach, which comprises the use of cultural, morphological, physiological and 85 pathogenicity characters in combination with phylogenetic analysis of nucleic acid 86 sequences. 87 The earliest phylogenetic studies of Colletotrichum using DNA sequences 88 were published by Mills et al. (1992) and Sreenivasaprasad et al. (1992). 89 Polymorphisms in the ITS1 region of the nrDNA were used to distinguish 90 Colletotrichum species. However, while the nrITS region is the most widely 91 sequenced region and has been chosen as the barcode locus for the Fungi, the utility 92 of this region is limited for systematic studies in Colletotrichum. Species diversity is 93 usually underestimated when based on nrITS sequences alone (Crouch et al., 94 2009a) and it has been demonstrated to have little phylogenetic utility (Doyle et al. 95 2013; Vieira et al. 2017). However, several additional markers have been applied for 96 multilocus phylogenetic inference to resolve the boundaries of cryptic species in the bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 97 genus (Cai et al. 2009; Damm et al. 2009; Doyle et al., 2013; Hyde et al., 2009; Lima 98 et al., 2013; Liu et al., 2016; Samarakoon et al., 2018; Veloso et al., 2018; Vieira et 99 al., 2014, 2017). 100 According to the most recent synopsis of the genus published in 2017, 188 101 Colletotrichum species have been described and incorporated into molecular 102 phylogenies. Among these species, 164 were distributed among 11 species 103 complexes and an additional 24 species had not been assigned to a species complex 104 (Marin-Felix et al., 2017). Additional species were recently described and three 105 additional clades were declared to represent new species complexes (Cao et al., 106 2018; Damm et al., 2019; Samarakoon et al., 2018). Due to its global distribution and 107 ecological and economic importance, research groups around the world are working 108 concomitantly to address regional diversity. However, the set of phylogenetic 109 markers used to discriminate species is variable by species complex and no standard 110 set of markers has been adopted based on objective criteria (Marin-Felix et al., 111 2017), making it difficult to combine data from disparate research groups and reliably 112 infer phylogenies and delimit species boundaries. Currently, thirteen different 113 molecular markers are commonly sequenced among the various Colletotrichum 114 species complexes: actin (ACT), DNA lyase (APN2), intergenic spacer between DNA 115 lyase and the mating-type locus MAT1-2-1 (APN2/MAT-IGS), calmodulin (CAL), 116 chitin synthase (CHS-1), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 117 intergenic spacer between GAPDH and a hypothetical protein (GAP2-IGS), 118 glutamine synthetase (GS), histone 3 (HIS3), nuclear rDNA internal transcribed 119 spacers (nrITS), mating type gene (MAT1-2-1), manganese-superoxide dismutase 120 (SOD2), and β-tubulin (TUB2). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 121 It is known that the efficiency of PCR amplification and the distribution of 122 phylogenetic informativeness of a given marker varies among species complexes 123 (Hyde et al., 2013). Most studies on the utility and reliability of individual markers 124 come from the Colletotrichum gloeosporioides complex (Cai et al., 2009; Liu et al., 125 2015; Sharma et al. 2013, Silva et al., 2012;2015; Vieira et al. 2017), and while the 126 C. gloeosporioides complex has been exhaustively studied in recent years, 127 recommendations on marker choice seem to be largely ignored. A classic case is the 128 utility of APN2/MAT-IGS, the most powerful marker to discriminate species within the 129 C. gloeosporioides complex: several recent studies described novel species within C. 130 gloeosporioides complex while excluding data from APN2/MAT-IGS (Costa et al., 131 2018; Diao et al. 2017; Fu et al., 2019; Jayawardena et al., 2016; Oliveira et al., 132 2018; Sharma et al., 2017; Silva et al., 2017; Sousa et al., 2018; Wang et al., 2019). 133 As mentioned above, this limits our ability to combine data from regional studies to 134 develop an accurate understanding of global diversity and phylogenetic relationships 135 within the genus. 136 While robust phylogenetic inference and reliable species delimitation relies on 137 the use of quality markers, studies on the performance of different molecular markers 138 for phylogenetic inference are missing for the majority of Colletotrichum species 139 complexes. In addition to the challenges discussed above, this also presents 140 practical problems for plant pathologists and ecologists who are looking to reliably 141 identify a large collection of isolates using molecular data. It is impractical for 142 researchers to sequence several loci, many of which may be of little phylogenetic 143 utility, across several hundred isolates simply for species identification. The aim of 144 the present study was to evaluate the phylogenetic informativeness of different 145 molecular markers used in Colletotrichum systematics and determine the optimal set bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 146 of markers for each species complex. From this, we hope to establish a consensus 147 on the minimal set of markers that can be used for Colletotrichum species 148 identification and delimitation and provide a practical reference for the large 149 community of researchers working on developing a better understanding of global 150 diversity, life history, and ecology of the genus. 152 2. MATERIAL AND METHODS 153 2.1 Datasets 154 We compiled several datasets to analyze the barcoding utility, phylogenetic 155 signal, and genealogical concordance of ACT, APN2, APN2/MAT-IGS, CAL, CHS-1, 156 GAPDH, GAP2-IGS, GS, HIS3, nrITS, and TUB2 for all Colletotrichum species 157 complexes described to date except for the C. caudatum complex, which was 158 excluded because nrITS was the only marker available for all species. These 159 datasets were compiled from published sequences retrieved from GenBank 160 (Supplementary File S1). Since testing the accuracy of prior species delimitations 161 were not the main focus of this study, we assumed that species boundaries 162 established in previous studies were accurate. 163 Thirteen species complexes were investigated in our study: Colletotrichum 164 acutatum, C. boninense, C. dematium, C. destructivum, C. dracaenophilum, C. 165 gigasporum, C. gloeosporioides, C. graminicola, C. magnum, C. orbiculare, C. 166 orchidearum, C. spaethianum and C. truncatum. Some species within each complex 167 were not included in the alignments due to the absence of sequences for several 168 markers since some of the analyses employed in the present work do not allow 169 missing data. Some markers were not analyzed due to a small number of species or 170 isolates with sequences available. The inclusion of these markers will drastically bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 171 reduce the number of species that can be included in each set of analyses (e.g. 172 GAPDH, HIS3, APN2 and APN2/MAT-IGS in the C. graminicola species complex). 174 2.2 Multiple sequence alignment 175 Multiple sequence alignments (MSA) of each locus were estimated individually 176 for each species complex. Sequences were compiled using the GenBank tool 177 implemented in MEGA 7 (Kumar et al., 2016). MSAs were estimated with the online 178 version of MAFFT 7 (Katoh et al., 2002; Katoh & Standley, 2013) using the G-INS-i 179 iterative refinement method and the 200PAM / k=2 nucleotide scoring matrix. 180 External gaps were trimmed in MEGA 7 before uploading MSAs to the GUIDANCE2 181 server (http://guidance.tau.ac.il/ver2/) (Sela et al., 2015) to access the alignment 182 confidence scores under the following parameters: MAFFT as the MSA algorithm; 183 max-iterate=0; pairwise alignment method=6mer; 100 bootstrap replicates. Unreliable 184 alignment regions were filtered by masking residues with scores below the lowest 185 cutoff, as proposed by Vieira et al. (2017). MSAs were converted to nexus format, 186 concatenated, and partitioned into a multilocus matrix using SequenceMatrix 1.8 187 (Vaidya et al., 2011). The number of invariable (I), variable (V), singletons (S) and 188 parsimony informative (PI) characters of the single locus alignments were calculated 189 using DnaSP 5.10 (Librado and Rozas,2009). 191 2.3 DNA barcoding 192 The effectiveness of markers to discriminate species within each species 193 complex was assessed by the barcode gap distance and intra/inter-specific distance 194 overlap (Hebert et al. 2003). Intra- and inter-specific distances were calculated for 195 each single locus alignment in MEGA 7. Single isolate species were removed from bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 196 alignments. Distances were calculated under the Kimura-2-parameter model, 197 allowing for substitution rates to differ among transitions and transversions, uniform 198 rates among sites, and gaps treated as pairwise deletions. Distance values were 199 sorted in Microsoft Excel Professional Plus 2016 and summary statistics were 200 calculated (maximum, minimum and mean distance). The barcode gap was 201 represented by the difference between the mean interspecific and intraspecific 202 distances (Hebert et al. 2003). The distance overlap percentage represents how 203 much the intraspecific distance overlaps with the interspecific distance and was 204 calculated as follows: max intraspecific distance ÷ max interspecific distance × 100. 205 Markers useful as barcodes will have a large barcode gap and a small intra/inter- 206 specific distance overlap. 208 2.4 Assessment of phylogenetic informativeness 209 The phylogenetic informativeness of markers commonly employed in 210 Colletotrichum systematics was estimated using the application PHYDESIGN 211 (Lopez-Giraldez and Townsend, 2011). Maximum likelihood (ML) trees were inferred 212 for each species complex using the concatenated alignments reduced to a single 213 representative isolate per species. Phylogenies were estimated in RAxML - HPC2 214 (Stawatakis, 2014) implemented on CIPRES Science Gateway portal 215 (https://www.phylo.org/portal2/home.action). ML tree searches were done assuming 216 the GTRGAMMA model and bootstrap support calculated with 1000 pseudoreplicates 217 (-m GTRGAMMA -p 12345 -k -f a -N 1000 -x 12345). ML trees were converted to 218 rooted ultrametric trees using the ‘chronos’ function in the ape package (Paradis et 219 al., 2004) using R Studio 1.1.442 (R Core Team, 2017). Trees were calibrated with 220 an arbitrary time scale with time=0 at the tips and time=1 at the root. The ultrametric bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 221 trees and the corresponding partitioned alignment were used as input files in 222 PHYDESIGN and the substitution rates were calculated using the program HyPhy 223 (Pond et al., 2005). The substitution rates estimated by the maximum likelihood 224 algorithm used by HyPhy are nonsensical for some sites in the alignment resulting in 225 very recent ‘phantom’ peaks that have no biological meaning. These peaks are likely 226 the result of indels or ambiguous sites in the alignment, therefore these alignment 227 positions with poorly estimated substitution rates were excluded from some genes 228 prior to phylogenetic informativeness profiling at the recommendation of the authors 229 of PhyDesign (http://phydesign.townsend.yale.edu/faq.html). Since the markers 230 included in this study and in most systematic studies of Colletotrichum do not require 231 more than two sequencing reads to sequence the full length of the locus 232 (representing the same sequencing effort and cost), the phylogenetic informativeness 233 values (PIV) were calculated on a net basis. The variable PImax represents the time 234 in which a given marker reaches the maximum PIV and was used to determine the 235 divergence time in which the marker is most informative (Fong and Fujita, 2011). We 236 ranked the markers according to the PIV values and the usefulness of markers was 237 assessed through the profile shape: low and flat curves represent the least 238 informative markers; high and sharp peaks represent the most informative markers. 239 The percentage of markers that resolve a given species was estimated using a 240 Bayesian Concordance Analysis – BCA (Ané et al., 2007; Larget et al., 2010). 241 Although BCA is a coalescent-based method to estimate species trees (Ané et al., 242 2007; Baum, 2007; Larget et al., 2010), this methodology can also be used to 243 quantify the proportion of markers that support a given clade, which is represented by 244 the concordance factor (CF). Individual locus trees were inferred in MrBayes 3.2.6 245 (Ronquist et al., 2012) implemented on the CIPRES cluster with four runs, each run bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 246 with four Markov chain Monte Carlo (MCMC) chains run for 10,000,000 generations, 247 sampled every 5,000 generations, totaling 2,001 trees per run. The frequency of 248 distinct topologies in the posterior distribution were summarized using mbsum 249 distributed with BUCKy 1.4.4 (Ané et al., 2007; Larget et al., 2010) skipping the first 250 25% of the trees as burn-in (-n 501). Primary concordance trees were estimated 251 using bucky and tree summary files output by mbsum were used as input files. 252 Concordance analyses were performed with a discordance factor (α) set at 1, four 253 MCMC chains, 1,000,000 generations, and the first 25% generations were discarded 254 as burn-in (-a 1 -k 4 -n 1000000 -c 4 -s1 23546 -s2 4564). 255 The Genealogical Sorting Index (GSI) was employed to identify the markers 256 that recover each species as monophyletic. GSI is an objective method that infers the 257 proportion of input trees for which a clade (species as applied here) are found to be 258 monophyletic and if the observed monophyly is greater than would be observed by 259 chance given the size of the data matrix (GSI=1 indicates monophyly) (Cummings et 260 al., 2008; Sakalidis et al., 2011), and can also be used to compare individual markers 261 according to their ability to discriminate species (Doyle et al., 2013). This 262 methodology can be applied to phylogenies inferred from a single locus as well as 263 multilocus analysis (Sakalidis et al., 2011). ML analyses were performed using the 264 single and multi-locus concatenated alignments. Species with a single isolate were 265 removed from the alignments. Analyses were carried out in RAxML as described 266 above with the number of pseudoreplicates reduced to 100 and outgroup isolates 267 specified prior to analysis. Rooted bootstrap trees were used as input files and each 268 tip was assigned to a species. GSIs were calculated using the GSI.py script and P- 269 value estimated from 100 permutations of each dataset (Cummings et al., 2008). The bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 270 GSI values were converted to heatmaps in the Heatmapper web server 271 (http://www.heatmapper.ca) to aid in visualization (Babicki et al., 2016). 273 2.5 Selection of best minimal sets of markers 274 The selection of a minimal set of optimal markers for robust phylogenetic 275 inference within each Colletotrichum species complex was based on results from the 276 phylogenetic informativeness profiling and GSIs. The markers were selected 277 according to the following criteria: 278 1) A minimum of three markers per complex based on ranking according to 279 PIV were selected. Three independent markers allow for the application of 280 the genealogical concordance phylogenetic species recognition criteria – 281 GCPSR (Dettman et al., 2003, Taylor et al. 2000), a commonly applied set 282 of criteria for phylogenetic species recognition in fungal systematics. 283 2) All species must be recognized as monophyletic by at least one of the 284 selected markers. GSIs were checked to confirm if each species in the 285 complex is recovered as monophyletic (GSI = 1) by at least one of the 286 selected markers. 287 3) ML trees were inferred from concatenated alignments of the three best 288 markers; if some species clade was poorly supported and/or all species 289 were not recovered as in the multilocus analyses with all markers 290 (unresolved relationships/polytomy), other markers were progressively 291 concatenated in decreasing order of phylogenetic informativeness until all 292 species were well resolved. 293 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 294 Once the best markers were chosen, the GSI was calculated for this 295 phylogeny to determine the level of species monophyly when only the best markers 296 are concatenated. The BCAs were also performed with this dataset to elucidate if the 297 species could be recognized by the majority of the selected markers. 299 3. RESULTS AND DISCUSSION 300 3.1 Alignment statistics 301 GAPDH, HIS3, and TUB2 were the most variable markers in the majority of 302 Colletotrichum species complexes, with PI characters ranging from 10—109, 11—82, 303 and 12—114, respectively (Table 1). PI characters for GS ranged from 63 to 93 and it 304 was the most variable marker within the C. gigasporum and C. orbiculare species 305 complexes. The APN2/MAT-IGS and GAP2-IGS, which are employed only in the C. 306 gloeosporioides species complex, had 192 and 115 PI characters, respectively, and 307 were the most variable markers within this complex. In contrast, nrITS presented the 308 fewest PI characters for most species complexes (0—36), followed by CHS-1 (3—45) 309 and ACT (3—63). 310 Most markers employed for Colletotrichum systematics comprise partial 311 sequences of orthologous protein-coding genes. These markers are composed of 312 introns flanked by long exons that are highly conserved. The GAPDH, GS, HIS3 and 313 TUB2 markers contain long intronic regions and APN2/MAT-IGS and GAP2-IGS 314 present long intergenic regions. This may explain why these markers are more 315 variable than the protein-coding loci. While protein-coding loci may be useful for 316 providing support along the backbone of the phylogeny within a species complex or 317 across the genus, markers with variable introns and intergenic sequences are 318 preferable for application at lower taxonomic levels (Schmitt et al., 2009). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 320 3.2 DNA barcoding feasibility 321 GAPDH had the largest barcode gap distance in seven of the 11 species 322 complexes evaluated. The percentage overlap between intra- and interspecific 323 distances was less than 20%, with the exception of the C. gigasporum and the C. 324 gloeosporioides species complexes (28.6% and 29.2%, respectively). GS has the 325 highest barcode distance with the lowest overlap for the C. gigasporum (0.12 and 326 12.1%) and C. orbiculare (0.08 and 2.7%) complexes. APN2/MAT-IGS had the 327 largest barcode gap distance (0.15) and the smallest overlap percentage (3.26%) 328 within the C. gloeosporioides species complex, making it the best candidate barcode 329 locus for the complex HIS3 had the largest barcode gap (0.026) within the C. 330 orchidearum complex with a relatively low overlap (16.9%), although CHS-1 had a 331 comparable barcode gap (0.025) with only a slightly higher overlap (21.6%). While 332 other markers, such as HIS3 and TUB2, are good candidate secondary barcode 333 markers in several complexes, nrITS was universally the poorest barcode candidate 334 with the lowest barcode gap distance within all species complexes. 335 Our results demonstrate that selecting a universal barcode marker for all 336 Colletotrichum species complexes among the markers currently being used is not 337 possible. An illustration of this is GAPDH. This marker is the best candidate barcode 338 marker for the majority of complexes, however it is among the worst barcode 339 candidates for the C. gloeosporioides species complex. While our results are not in 340 agreement with Cai et al. (2009), which carried out the first study evaluating the 341 markers to discriminate species within C. gloeosporioides species complex, only five 342 species were included in their analyses and the Musae and Kahawae clades (sensu 343 Weir et al. (2012)) were treated as single species. Moreover, while GAPDH was bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 344 chosen as the best marker relative to EF1α, ACT, CHS-1 and nrITS, the latter three 345 markers perform very poorly for species delimitation in the C. gloeosporioides 346 complex (Vieira et al., 2017) and, therefore, we did not include them in any of our 347 analyses. While GAPDH, with its large barcode gap and small overlap along with the 348 ease with which it can be amplified and sequenced, makes it among the best 349 barcode candidates across Colletotrichum as a whole, the selection of the best 350 barcode markers is dependent on the species complex. 351 The search for a universal barcode locus for the genus will require a 352 comparative analysis across the genomes of several species in all species 353 complexes. Intergenic sequences in syntenic regions of the genome are good 354 candidates if APN2/MAT-IGS and GAP-IGS are any indication. Intergenic sequences 355 may provide fast-evolving phylogenetic markers to be used for population genetic 356 and phylogenetic studies on fungal species complexes (Magain et al., 2017). 358 3.3 Optimal markers for phylogenetic inference 359 The most informative markers differ among the species complexes within 360 Colletotrichum. Net phylogenetic informativeness profiles and their respective 361 ultrametric trees are presented in Fig. 2. PIV and PImax values are summarized in 362 Table 1. The GSI analyses illustrate the ability of a given marker to recover species 363 as monophyletic is dependent on the species complex (Fig. 3, GSI values presented 364 in Supplementary File S2). Most species that were monophyletic in the multilocus 365 tree with all markers (GSI near to 1) were also recovered as monophyletic when only 366 the most informative markers were concatenated. In parallel, the BCAs revealed that 367 the proportion of markers supporting individual species-level clades (expressed as bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 368 concordance factors) increase when the less informative markers are removed from 369 the analyses. 370 Since the suite of markers differ by species complex and the performance of a 371 given marker differs among the species complexes, the results of the phylogenetic 372 informativeness profiling, GSI, Bayesian concordance analyses and selection of the 373 minimum set of markers for robust phylogenetic inference are summarized below for 374 each species complex. 376 3.3.1 Colletotrichum acutatum s. l. 377 HIS3, GAPDH and TUB2 are the most phylogenetically informative markers 378 within the C. acutatum species complex, with PImax at 0.67, 0.78 and 0.99, 379 respectively (Fig. 2). All markers currently used for systematics of the C. acutatum 380 complex have an optimum inferential timescale varying from 0.67 to 0.99, which is 381 useful to resolve deeper relationships. However, the absence of markers with lower 382 PImax negatively impacts the our ability to infer relationships among recently 383 diverged species. This is evident by the low internal node support within Clade 1 and 384 Clade 2 sensu Damm et al. (2012a), which are the most recently diverged lineages 385 within the complex (Supplementary File S3). Species are strongly supported in the 386 concatenated multilocus analysis (ML support ≥ 70%, Supplementary File S3), 387 whereas the relationship among them are poorly resolved even when only the best 388 markers were used to build the tree. In contrast, nodes throughout Clade 5 are 389 strongly supported. The inclusion of a marker with PImax near to 0.2 could improve 390 the support on the deeper nodes of the Clades 1 and 2. 391 Most species in C. acutatum s. l. could be resolved by the majority of the 392 markers only when the optimal markers are combined (Fig. 4B). Seven (C. austral, C. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 393 chrysanthemi, C. fioriniae, C. johnstonii, C. lupini, C. nymphaeae and C. tamarilloi) 394 among the 17 multiple-isolate species were supported by the majority of individual 395 genes (CF≥0.66) in the BCA when all six loci were used in the analysis (Fig. 4A), 396 which means that four out of six markers support the monophyly of these species. 397 The CF increases when the analyses included only HIS3, GAPDH and TUB2, and 398 only four (C. cosmi, C. costaricense, C. paranaense and C. phormii) out of 17 399 multiple-isolate species presented CF<0.66 (Fig. 4B). The GSI (Fig. 3) indicates that 400 C. melonis is recovered by GAPDH and C. phormii by HIS3. The GSI for 401 Colletotrichum costaricense was nearly 1 for both GAPDH and TUB2 (0.95 and 0.99, 402 respectively) and less than 0.5 for all other markers. This is consistent with a CF of 403 0.63, indicating that nearly 2 of the 3 loci included in the analysis support the 404 monophyly of the species. Colletotrichum paranaense had a low GSI for most 405 markers and CF=0.22, which means that no individual marker fully supports this 406 species as monophyletic. However, C. paranaense is strongly supported in the 407 multilocus concatenated analysis and has a high GSI when analyses included all or 408 the best markers. These results clearly show that the multilocus trees masks the 409 incongruences among the individual gene trees or the use of markers with low 410 phylogenetic signal. 411 Colletotrichum paranaense was described by Bragança et al. (2016) as a 412 species closely related to C. limetiicola and C. melonis (Damm et al., 2012a). 413 Phylogenetic species recognition was not employed in the delimitation of C. 414 paranaense. The species was recognized based on the topology of the multilocus 415 tree and differed from C. limetticola and C. melonis by percentage identity according 416 to blastn searches. Our results suggest that the relationship among these species 417 and the delimitation of C. paranaense needs to be revisited. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 419 3.3.2 Colletotrichum boninense s. l. 420 CAL, TUB2 and GAPDH were the most informative markers to resolve species 421 within C. boninense s. l. (Fig. 2). All three markers are comparable in levels of 422 phylogenetic informativeness, with PIV ranging from 47 to 53. However, these 423 markers differ with respect to the optimal timescale at which they are informative. 424 CAL and TUB2 peaked at 0.58 and 0.63, respectively, providing more signal towards 425 the root of the phylogeny. In contrast, GAPDH reached maximum informativeness at 426 0.24, which is more useful for resolving species in this complex, since the divergence 427 epoch for most species is near 0.2. 428 Most species can be resolved by most of the markers according to the 429 concordance analyses that include all markers and only the best markers (CF>0.57 430 and >0.66, respectively) (Fig. 5B). GAPDH supports most species as monophyletic 431 (GSI 0.98—1) (Fig. 2), with the exception of three species (C. catinaense, C. 432 constrictum and C. limonicola). According to Damm et al. (2012b), all the species 433 within the C. boninense species complex can be identified by sequencing GAPDH. In 434 contrast, our results show that C. catinaenese, C. constrictum and C. limonicola are 435 weakly supported as monophyletic by GAPDH (GSI = 0.51, 0.69 and 0.59, 436 respectively) and could only be recovered as monophyletic in analyses of TUB2, HIS 437 or ACT. Colletotrichum cymbidiicola and C. novae-zelandiae were recovered as 438 monophyletic with GAPDH and C. limonicola by ACT and TUB2 (Fig. 3). Thus, 439 concordance factors for these species remained low even when the best markers 440 were used in the BCA. Additionally, C. novae-zelandiae was the only species for 441 which GSI<1 in both multilocus analyses (Fig. 3). Colletotrichum catinaense could be 442 resolved by two of the best markers (CAL and TUB) and by ACT, which led the CF to bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 443 shift from 0.56 in the BCA with all markers to 0.73 when only the best markers were 444 included (Fig. 5). The use of CAL, TUB2 and GAPDH leads to a slight increase in 445 support across the multilocus ML tree when compared with the tree reconstructed 446 with the whole dataset (Supplementary File S4), with the exception of Colletotrichum 447 novae-zelandiae, which remained poorly supported. 449 3.3.3 Colletotrichum dematium s. l. 450 GAPDH, HIS3 and TUB2 presented high and sharp profiles (Fig. 2), with 451 PImax 0.05—0.06 and were the most useful markers to discriminate species within 452 the C. dematium species complex. HIS3 was the only marker able to discriminate all 453 species in the complex. nrITS, CHS-1 and ACT presented low and flat curves and 454 were the least informative markers (Fig. 2). Colletotrichum anthrisci and C. circinans 455 could be resolved by all markers (CF=1, Fig. 6), while Colletotrichum spinaceae was 456 resolved as monophyletic by all markers except nrITS. The exclusion of nrITS in the 457 BCA increased the CF from 0.86 to 1. The relationship between Colletotrichum 458 dematium and C. lineola was not clearly resolved in the multilocus analyses. 459 Although C. dematium isolates were well supported as monophyletic in the tree 460 inferred from all markers, C. lineola isolates remained paraphyletic (Supplementary 461 File S5). When only GAPDH, HIS3 and TUB2 were considered, the isolates of C. 462 dematium and C. lineola were placed together in a poorly supported clade. 463 Colletotrichum lineola was recovered as monophyletic only by HIS3 (GSI=1). Both 464 ACT and GAPDH could also resolve this species, albeit with low GSI (0.83 and 0.73 465 respectively). In contrast, these three markers recover C. dematium with high level of 466 monophyly (GSI 0.97—1). Colletotrichum lineola presented low monophyly level in all bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 467 multilocus GSI analysis and C. dematium was not monophyletic when only the best 468 markers were considered. 469 In the Damm et al. (2009) study, the isolates of C. dematium and C. lineola are 470 separated into two short-branch clades. They chose to retain both taxa, but raised 471 the hypothesis that C. dematium and C. lineola are different populations within the 472 same species. We performed the GSI analyses with the multilocus trees combining 473 C. dematium and C. lineola in the same group (data not shown). The group was 474 recovered as monophyletic when all markers and only the best markers are 475 concatenated (GSI=1 and 0.99 respectively), which support the hypothesis that C. 476 dematium and C. lineola are the same species. Colletotrichum eryngiicola, C. 477 hemerocallidis, C. insertae, C. quiquefoliae and C. sonchicola were placed together 478 with C. dematium and C. lineola in a polytomous clade in Samarakoon et al. (2018) 479 study, which indicates that those species may also be members of the group C. 480 dematium/ C. lineola. These species were not included in our analyses due to 481 absence of HIS3 and TUB2 sequences. In the future, species boundaries within this 482 lineage need to be revisited. 484 3.3.4 Colletotrichum destructivum s. l. 485 HIS3, TUB2 and GAPDH were the most informative markers to resolve 486 species within C. destructivum complex (Fig. 2). GAPDH and HIS3 possess the 487 phylogenetic signal to resolve shallow clades (PImax=0.28 and 0.37 respectively), 488 whereas TUB2 performs better in the deep branches (PImax=0.73). HIS3 was the 489 most informative marker and was able to six (C. destructivum, C. lentis, C. lini, C. 490 tabacum, C. utrechtense and C. vignae) out of the nine species analyzed as 491 monophyletic (GSI 0.98—1) (Fig. 3). Although GAPDH is one of the best markers, bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 492 only four species (C. fuscum, C. lentis, C. tabacum and C. vignae) were highly 493 monophyletic in the topologies provided by this locus (GSI 0.9—1) (Fig. 3). 494 Colletotrichum americae-borealis was not monophyletic in the topology of any single 495 or multi-locus trees (GSI 0.34—0.89). Four (C. destructivum, C. lentis, C. tabacum, 496 C. utrechtense) out of nine multiple-isolates species could be recovered by most of 497 the genes in the concordance tree (CF≥0.66) inferred from all markers (Fig. 7A). 498 These same four species plus C. vignae were recovered by most of the genes 499 (CF≥0.66) when only the optimal markers were considered. Moreover, the CF and 500 the bootstrap supports of the internal nodes increased when only the best markers 501 were combined (Fig. 7 and Supplementary File S6). 502 The clade comprised by C. americae-borealis and C. lini forms a polytomy 503 when only the three best markers were used (Supplementary File S6). We also 504 tested the inclusion of ACT or nrITS in the concatenated dataset. However, this clade 505 remains a polytomy and can only be resolved when all markers are concatenated. 506 The overall GSI values also reduced when ACT or nrITS was included in the 507 multilocus analysis (data not shown). Thus, it is not reasonable to include ACT and 508 nrITS in the analysis due to their low phylogenetic informativeness. Other markers 509 with greater phylogenetic signal need to be tested for this complex in order to resolve 510 the relationships among the unresolved clades and clarify the species identities. 512 3.3.5 Colletotrichum dracaenophilum s. l. 513 TUB2, GAPDH and HIS3 presented the highest PIV (58, 52 and 36 514 respectively) and were the most informative markers to distinguish species within the 515 C. dracaenophilum complex (Fig. 2). All the markers peaked around the same 516 timescale (PImax 0.39—0.46) and provide robust support for relationships in both bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 517 deep and shallow nodes. Although TUB2, GAPDH and HIS3 were the most 518 informative markers, all the markers are informative enough to discriminate species 519 in C. dracaenophilum s. l. (GSI 0.99—1) (Fig. 3). This is corroborated by the high 520 concordance factors (CF=1) in BCAs (Fig. 8) and 100% support in multilocus 521 analyses (Supplementary File S7) when all or the best markers were combined. 523 3.3.6 Colletotrichum gigasporum s. l. 524 GS, CAL and GAPDH were the most powerful markers to discriminate species 525 within C. gigasporum s. l. (Fig. 2), with PImax at 0.1, 0.44 and 0.28 respectively. 526 Although GS, CAL and GAPDH were the most informative markers, all the markers 527 and the concatenated datasets were able to discriminate all species within the C. 528 gigasporum complex (GSI 0.99—1) (Fig. 3), as reported by Liu et al. (2014). All 529 species presented maximum CF (Fig. 9) and ML support (Supplmentary File S8) 530 independently of which set of markers was included in these analyses. 532 3.3.7 Colletotrichum gloeosporioides s. l. 533 APN2/MAT-IGS, GAP2-IGS and APN2 were the most informative markers to 534 separate species within the C. gloeosporioides species complex (Fig. 2). The 535 informativeness profiles indicate APN2/MAT-IGS and APN2 are of peak 536 informativeness at 0.19 and 0.17, respectively, and are informative for shallow 537 divergences, whereas GAP2-IGS has an optimal timescale at 0.42 and provides 538 more signal for resolving deep nodes. This range in values for peak informativeness 539 led to high support at both deep and shallow nodes when only the most informative 540 markers were used for phylogenetic inference (Supplementaray File S9). APN2/MAT- 541 IGS could separate all the species included in the analyses, although only C. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 542 fragariae, C. queenslandicum, C. siamense and C. tropicale reached a GSI of 1 (Fig. 543 3). Moreover, all species were recovered as monophyletic in all multilocus analyses. 544 Colletotrichum siamense was the only species that could not be recovered by most of 545 the markers, and its CF reduced from 0.59 in all-markers analysis to 0.38 in the best- 546 markers analysis (Fig. 10B). The CF reduced due to removing CAL and TUB2, which 547 also recover C. siamense as monophyletic (GSI=0.99). 548 ACT, CHS-1 and nrITS were not included in our analyses because these 549 markers were previously reported as the worst markers to distinguish species in C. 550 gloeosporioides complex (Vieira et al., 2017). In the present study, rather than 551 restating the results of Vieira et al. (2017), we evaluated if the seven markers 552 proposed by those authors are needed for diversity studies and species assignment. 553 We determined sequences of APN2/MAT-IGS, GAP2-IGS and APN2 are sufficient to 554 resolve the species in C. gloeosporioides s. l.. However, sequences of GAP2-IGS 555 and APN2 are not available for all species within C. gloeosporioides s.l.. These data 556 should be generated for all species within the species complex. 557 Colletotrichum gloeosporioides s. l. is the most studied species complex in the 558 genus. More than 10 different markers have been used for systematics and 559 taxonomy of the C. gloeosporioides complex over the last 10 years. However, there 560 is no agreement about which markers should be used for species recognition. The 561 first attempt to determine the best markers was done by Cai et al. (2009), in which 562 GAPDH was the best marker to discriminate species in C. gloeosporioides s. l., and 563 the set composed by ACT, GAPDH, nrITS and TUB2 was recommended for 564 multilocus analysis. Weir et al. (2012) highlighted that although GAPDH is one of the 565 most effective markers to distinguish species within C. gloeosporioides complex, the 566 combination with GS is necessary to distinguish some species. Our study shows that bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 567 GAPDH was the least informative marker among the ones included in this study for 568 C. gloeosporioides s. l.. This marker was the least variable (Table 1), and had the 569 smallest barcode gap and the largest overlap distance (Fig. 1). Colletotrichum 570 fragariae was the only species recovered as monophyletic (GSI=1) by this marker 571 (Fig 2). Based on this, GAPDH is considered one of the worst barcode candidates for 572 the C. gloeosporioides complex among the markers tested in the present study. 573 More recently, several studies demonstrate the singular ability of APN2/MAT- 574 IGS to discriminate species in the C. gloeosporioides complex (Sharma et al., 2013, 575 2015; Vieira et al., 2014), although others previously demonstrated the utility of this 576 marker (Du et al., 2005; Rojas et al., 2012; Silva et al., 2012). The main issue in 577 using APN2/MAT-IGS was the splitting out of C. siamense in a species complex, in 578 which several monophyletic lineages could be revealed by the phylogeny inferred by 579 this marker (Sharma et al., 2013). Although the multilocus analysis had been done, 580 the lineages identity was confirmed only based on the APN2/MAT-IGS, since this 581 marker performs good as well as the multilocus matrix. Other studies use the same 582 criteria to discriminate species and several species within C. siamense s. l. were 583 described (Sharma et al., 2015; Vieira et al., 2014). Later, Liu et al. (2016) use 584 coalescent methods for phylogenetic species delimitation and synonymize all those 585 species in the complex into C. siamense. It was revealed incongruences among the 586 APN2/MAT-IGS tree and other individual gene trees. The study clarifies how 587 multilocus analyses can mask discordances among individual gene trees and lead to 588 species misidentification. 589 The combination of APN2/MAT-IGS and GS were proposed as the barcode to 590 delimit species within C. gloeosporioides complex (Liu et al., 2015). These two 591 markers were collectively powerful enough to discriminate the 22 species included in bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 592 the study and produced the same topology as that inferred from a 6 marker 593 multilocus dataset (ACT, CAL, GAPDH, GS, nrITS and TUB2). We tested the 594 combination of APN2/MAT-IGS with each remaining marker individually and all 595 resulting trees were similar in topology (data not shown). Some species, such as C. 596 siamense, are polyphyletic in our GAPDH and GS trees, which belies the 597 incongruence with the multilocus and the other individual gene trees as presented by 598 Silva et al. (2012) and Vieira et al. (2017). Mating-type associated markers, such as 599 APN2/MAT-IGS and MAT1-2, had fast evolutionary rates and high variability, and 600 may dominate the topology of multilocus trees (Liu et al., 2016). Since the 601 combination of APN2/MAT-IGS with any other marker produces similar topologies, 602 the combination with any other marker besides GAPDH and GS is preferable in order 603 to avoid inconsistencies in species delimitation within the C. gloeosporioides 604 complex. While we find that APN2 is another informative marker that should be 605 incorporated into diversity studies of the C. gloeosporioides complex, the proximity of 606 this marker to the APN2/MAT-IGS region suggests it is part of the same linkage 607 group and thus may not represent an independent sample of an organisms 608 evolutionary history. If this is the case, linking substitution models and inferred trees 609 across these two loci would be required and the addition of another locus may be 610 necessary to fulfill the expectations under GCPSR. 611 Although the majority of Colletotrichum species belong to the C. 612 gloeosporioides complex, identification and description of taxa within this complex is 613 the most problematic. Sequences of the best markers, mainly APN2 and GAP2-IGS, 614 have not been sequenced for the majority of species, which prohibits the detection of 615 these species using these markers. A priority should be to generate sequences of the 616 best markers for all species to avoid further misidentification and the introduction of bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 617 dubious taxa. Until this happens, the description of new species within the C. 618 gloeosporioides complex will likely require sequencing multiple additional markers to 619 be certain of their novelty. 621 3.3.8 Colletotrichum graminicola s. l. 622 TUB2 was the most informative marker, followed by nrITS and ACT (Fig. 2). In 623 contrast, CHS-1 was ranked as the worst marker for the C. graminicola species 624 complex. GSI and BCA analyses could not be performed for this complex because 625 only one isolate per species could be included in the analyses. Other markers such 626 as APN2, APN2/MAT-IGS, GAPDH, HIS3, MAT1-2 and SOD2 were also used to 627 identify species within C. graminicola complex (Cannon et al., 2012; Crouch et al. 628 2009b, c; Crouch and Tomaso-Peterson, 2012; Du et al., 2005; Moriwaki and 629 Tsukiboshi, 2009; O'Connell et al., 2012; Tao et al., 2013). However, sequences for 630 these markers are missing for several species. According to our results for other 631 complexes, APN2, APN2/MAT-IGS, GAPDH and HIS3 are likely to be more powerful 632 markers than those that we could include in the analyses for the C. graminicola 633 complex. Sequences for these markers need to be generated for isolates of several 634 species in order to establish a better set of optimal markers for phylogenetic 635 inference and species discrimination in the C. graminicola complex. 637 3.3.9 Colletotrichum magnum s. l. 638 HIS3, TUB2 and GAPDH were the most informative markers to discriminate 639 species within the C. magnum complex (Fig. 2). Although HIS3 presented the highest 640 PIV (26), TUB2 presented a higher PImax (0.52 versus 0.44) and is able to separate 641 more species than HIS3. Only C. brevisporum and C. magnum were accounted for bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 642 by more than one isolate and could be analyzed in the BCA and GSI analyses (Fig. 643 3). Colletotrichum brevisporum reach high levels of monophyly only in analyses of 644 the concatenated dataset (GSI 0.98—1, Fig. 3) and only one gene can fully resolve 645 this clade (CF=0.4, Fig 11). In contrast, C. magnum is recovered as monophyletic by 646 the multilocus dataset and by the best markers. However, the CF reached 1 only 647 when the best markers were concatenated (Fig 11). Colletotrichum magnum 648 remained highly supported when only the best markers were used to reconstruct the 649 phylogeny of the C. magnum complex (Supplementary File S10B) with only a slight 650 decrease in support at some internal nodes. The species C. liaoningense was not 651 included in the analyses due to sequence deposition errors detected by Damm et al. 652 (2019) from the study were C. liaoningense was described (Diao et al., 2017). Damm 653 et al. (2019) concluded that C. liaoningense needs to be revisited. 655 3.3.10 Colletotrichum orbiculare s. l. 656 GS stood out from all markers as the most informative (PIV=69) in the C. 657 orbiculare complex, followed by HIS3 and GAPDH (PIV=37 and 32 respectively) (Fig. 658 2). GS peaks at 0.53 and can discriminate the majority of species and provide robust 659 support for the relationships among them, since the divergence time for most species 660 is about 0.3. On the other hand, HIS3 and GAPDH are useful to discriminate some 661 recently diverged species (PImax=0.13 and 0.17 respectively) such as C. orbiculare 662 × C. sidae and C. trifolii × C. malvarum. GS was the marker that recovered more 663 species as monophyletic (GSI=1), with the exception of C. sidae (GSI=0.89) (Fig. 3). 664 Colletotrichum lindemuthianum was the only species that was not recovered as 665 monophyletic when all or the best markers were concatenated (both GSI=0.78), 666 which is likely due to the variability within this species as currently circumscribed. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 667 Damm et al. (2013) split C. lindemutianum into two different lineages (C. 668 lindemuthianum 1 and 2), which are observed in the nrITS and GS trees. In our 669 multilocus trees, the isolates CBS133.57 and CBS131.57 where moved to the clade 670 C. lindemuthianum 1 and CBS150.28 to the clade C. lindemuthianum 2. This result is 671 discordant with that in Damm et al. (2013) and we conclude that the terminal clades 672 within C. lindemuthianum represent intraspecific variability. Thus, both C. 673 lindemuthianum lineages were not considered separate species in our analysis. The 674 CF of both deep and shallow nodes increases significantly when only the best 675 markers were analyzed (Fig.12). Reducing the set of markers does not cause 676 significant differences in the tree topology or clade support (Supplementary file S10). 678 3.3.11 Colletotrichum orchidearum s. l. 679 TUB2 and HIS3 were the most powerful genes capable of discriminating 680 species within the C. orchidearum complex (PIV=17 and 16 respectively), followed by 681 CHS-1 (PIV=12) (Fig. 2). TUB2 and HIS3 are good markers to discriminate all 682 species in the complex and support the relationships among them (PImax=0.8 and 683 0.67 respectively), whereas CHS-1 can help to distinguish and support recently 684 diverged species (PImax=0.47). All species were recovered with high levels of 685 monophyly with TUB2 (GSI=0.99—1), and also by both multilocus analyses 686 (GSI=0.97—1) (Fig. 3). Colletotrichum musicola was the only species that could be 687 recovered by all markers evaluated (CF=1) (Fig. 13). The CFs increased when the 688 less informative markers were removed from the analysis and most species were 689 recovered by all markers (CF=1), with the exception of C. cliviicola and C. plurivorum 690 (CF≥0.66). Damm et al. (2019) also reported that some clades were not supported by 691 some of the individual gene analyses. All deep and shallow nodes retained high bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 692 support when only the best markers were combined in our analyses (Supplementary 693 File S12B). 695 3.3.12 Colletotrichum spaethianum s. l. 696 GAPDH, TUB2 and ACT were the best markers to separate species within the 697 C. spaethianum complex (Fig. 2). These markers peak at approximately 0.4 (PImax 698 0.47—0.77), which is in the range of where most species in the complex diverge 699 indicating they can separate the majority of species in this complex. However, 700 GADPH is the only marker that recovers all species as monophyletic (GSI 0.98—1) 701 (Fig. 3). All species reached complete monophyly (GSI=1) in both multilocus trees. 702 Colletotrichum liriopes and C. spaethianum could be recovered by most of the genes 703 only when the best markers are considered in the BCA (CF=0.77 and 0.70, 704 respectively) (Fig. 14B). All species remain strongly supported when only the best 705 markers were concatenated (Supplementary Figure S13B). 706 HIS3 appears to be a good marker in several other Colletotrichum complexes 707 according to the present study, and may be among the three best markers for C. 708 spaethianum complex. However, sequences of HIS3 and other markers such as 709 CHS-1 and CAL are available only for one isolate of several species and we cannot 710 include them in our analyses. Sequences of these markers need to be generated for 711 other isolates to determine if one of these markers could be used to substitute for 712 ACT, which does not perform well for separating Colletotrichum species. 714 3.3.13 Colletotrichum truncatum s. l. 715 GAPDH was clearly the most informative marker (PIV=57), followed by TUB2 716 and ACT (PIV=33 and 23 respectively) (Fig. 2). All markers peaked above 0.4 and bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 717 were able to discriminate most species. Colletotrichum acidae and C. curcumae were 718 monophyletic (GSI=1) in all single and multilocus datasets (Fig. 3) with CFs equal to 719 1 in analyses of both all and best markers BCA (Fig. 15). These species were also 720 supported by maximum bootstrap values in the multilocus trees (Supplementary File 721 S14). Colletotrichum corchorum-capsularis and C. truncatum were not recovered as 722 monophyletic by any dataset. Moreover, C. truncatum is paraphyletic or polyphyletic 723 in the multilocus trees (Fig. 15, Supplementary File S14), which leads us to the 724 conclusion that C. corchorum-capsularis, as circumscribed by Niu et al. (2016), 725 cannot be recognized as a species distinct from C. truncatum. Our results strongly 726 suggest that C. corchorum-capsularis and C. truncatum may be the same species 727 and isolates of both species were placed together in a clade with high CF (≥0.96) in 728 the BCAs and maximum support in the multilocus analyses (Fig. 15, Supplementary 729 File S14). Additional work is needed using the best markers for the C. truncatum 730 complex and objective species recognition methods to determine the taxonomic 731 status species boundaries of C. corchorum-capsularis and C. truncatum. 733 4. Conclusions 734 We used phylogenetic informativeness profiling, maximum likelihood and 735 coalescent-based phylogenetic analyses, measures of barcode utility, and 736 genealogical sorting indices to assess the performance of the several molecular 737 markers used in Colletotrichum systematics and taxonomy across all known species 738 complexes. While HIS3, GAPDH, and TUB2 were among the best markers for most 739 of the complexes, the optimal set of markers is not always the same across all 740 complexes. ACT, CHS-1 and nrITS were the worst markers and, as previously 741 proposed for the C. gloeosporioides complex (Vieira et al., 2017), they can be bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 742 discarded from the phylogenetic analysis of almost all species complexes. ACT was 743 retained in the set of best markers for the C. graminicola, C. spaethianum and C. 744 truncatum complexes to achieve a minimum of three markers as proposed in the 745 methodology of the present study. However, few markers were included for these 746 complexes due to missing data, therefore additional markers need to be sequenced 747 and their performance evaluated. The analyses of C. caudatum, C. dematium, C. 748 graminicola and C. spaethianum complexes were the most impacted by the 749 excessive amount of missing data for the majority of the markers, which highlights 750 the importance of selecting a standard set of markers to delimit species. Similarly, 751 several isolates and/or species were not included in the marker analyses for C. 752 gloeosporioides s.l. due to selective data acquisition by different research groups. It 753 is not clear how the inclusion of sequences from these isolates might impact our 754 results. Sequences need to be generated for these markers and/or species to 755 provide more decisive results. We have also identified species complexes, which will 756 need to be revisited in the future, in which it appears species have been misidentified 757 (C. acutatum, C. dematium and C. truncatum). 758 Selecting the optimal markers to sequence for biodiversity studies on 759 Colletotrichum will impact Colletotrichum studies in a few ways. First, species 760 recognition will likely be more accurate and robust by avoiding the confounding effect 761 of including markers with low phylogenetic signal in the analyses. Secondly, 762 phylogenetic studies of Colletotrichum will become more economical, since 763 sequencing markers with low informativeness represents a low return on investment. 764 Finally, if research groups take guidance from this study, we are more likely to see a 765 consensus developed on the data acquired for phylogenetic studies on Colleotrichum 766 and we will be closer to a global assessment with combinable data. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 767 Researchers around the world continue to have an interest in documenting the 768 diversity of Colletotrichum species associated with economically important plant 769 species. However, this work is labor intensive and expensive because several 770 hundred isolates are typically screened and a paucity of distinctive morphological 771 characters necessitates DNA sequencing. The expense has been unnecessarily 772 compounded by the lack of an objective and comprehensive assessment of the utility 773 of existing markers for phylogenetic inference and species identification/delimitation 774 and the lack of a consensus on the markers to be used. We hope the results 775 presented here will help to address this problem. While the optimal markers differ by 776 species complex, our results provide some guidance on the most efficient path to 777 document and describe diversity within Colletotrichum. Our results also show that for 778 the accurate identification and delimitation of Colletotrichum species, a small set of 779 markers with strong phylogenetic signal is more suitable than a large set including 780 markers with both weak and strong phylogenetic signal. GAPDH is among the 781 optimal set of markers for 10 of the 13 species complexes in Colletotrichum, followed 782 by TUB2 (10 of 13), and HIS3 (7 of 13). Therefore, GAPDH is a good marker to 783 sequence for initial diversity screening and assigning isolates to a species complex 784 because data for this marker is available for the majority of species within the genus. 785 However, selection of additional markers for phylogenetic inference and species 786 delimitation will depend on the species complex. 787 Finally, while we recommend the optimal markers for species recognition 788 within Colletotrichum in order to improve diversity studies in the genus, our 789 understanding of evolutionary relationships among species remains poorly resolved. 790 Improving our understanding of relationships among taxa within Colletotrichum will 791 require more robust genomic sampling. Genome sequencing is underway for many bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 792 species of Colletotrichum, however a comprehensive phylogenomic study of the 793 genus is needed. 795 Acknowledgments 796 This study was funded by “Conselho Nacional de Desenvolvimento Científico 797 e Tecnológico – CNPq” (Universal number 408724/2018-8). Willie A. S. Vieira 798 acknowledges the “Fundação de Amparo à Ciência e Tecnologia do Estado de 799 Pernambuco – FACEPE” for the Postdoc fellowship (Process number BFP-0040- 800 5.01/16). Marcos P. S. Câmara acknowledge CNPq for the research productivity 801 fellows. Josiene S. Veloso acknowledge the “Coordenação de Aperfeiçoamento 802 Pessoal de Ensino Superior – CAPES” and the “Programa Nacional de Pós- 803 Doutorado/CAPES – PNPD/CAPES” for the Posdoc fellowship. This work represents 804 partial fulfillment of the degree of Master of Phytopathology for Priscila A. Bezerra. V. 805 P. Doyle would like to acknowledge funding support from the Louisiana Board of 806 Regents Research Competitiveness Subprogram (LEQSF(2016-19)-RD-A-01). 808 Glossary 809 Appressorium: specialized cell produced by some phytopathogenic fungi which is 810 used to infect plant hosts. 811 Conidium: asexual spore of Ascomycota and Basidiomycota. 812 Endophytic fungi: fungi that grow inside the plant tissues without causing disease 813 symptoms. 814 Phytopathogenic fungi: fungi that cause plant diseases. 815 Sensu lato (s. l.): taxonomic terminology used to reference species complexes (C. 816 acutatum species complex = C. acutatum s. l.). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 817 Sensu stricto (s. s.): when is necessary to refer the species with the same name of 818 the complex (C. acutatum s. s. is a species within C. acutatum s. l.). 819 Species complex: major clades strongly supported within Colletotrichum genus tree. 820 These clades include phylogenetic species closely related which most are 821 indistinguishable based on phenotypical characters (e.g. conidial and appressorial 822 shape and size, growth rate, color of colonies). Species complexes get the same 823 name of the species within them that is more known or that was firstly described. 824 In some cases, members within a given species complex share peculiar conidial 825 characteristics: C. acutatum – conidia with acute ends; C. boninense – presence of a 826 prominent scar (hilum) at the base of the conidium; C. caudatum – conidia with a 827 filiform appendage at the apex; C. gigasporum – longest and widest conidia within 828 the genus. 830 References 831 Ané, C., Larget, B., Baum, D.A., Smith, S.D., Rokas, A., 2007. Bayesian estimation 832 of concordance among gene trees. Molecular Biology and Evolution 24, 412– 833 426. 834 Babicki, S., Arndt, D., Marcu, A., Liang, Y., Grant, J.R., Maciejewski, A., Wishart, 835 D.S., 2016. Heatmapper: web-enabled heat mapping for all. Nucleic Acids 836 Research 44, W147–W153. 837 Baum, D.A., 2007. Concordance trees, concordance factors, and the exploration of 838 reticulate genealogy. Taxon 56, 417–426. 839 Bragança, C.A.D., Damm, U., Baroncelli, R., Massola Júnior, N.S., Crous, P.W., 840 2016. Species of the Colletotrichum acutatum complex associated with 841 anthracnose diseases of fruit in Brazil. Fungal Biology 120, 547–561. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 842 Cai, L., Hyde, K.D., Taylor, P.W.J., Weir, B.S., Waller, J., Abang, M.M., Zhang, J.Z., 843 Yang, Y.L., Phoulivong, S., Liu, Z.Y., Prihastuti, H., Shivas, R.G., McKenzie, 844 E.H.C., Johnston, P.R., 2009. A polyphasic approach for studying Colletotrichum. 845 Fungal Diversity 39, 183–124. 846 Cannon, P.F., Damm, U., Johnston, P., Weir, B.S., 2012. Colletotrichum – Current 847 status and future directions. Studies in Mycology 73, 181–213. 848 Cao, X.; Xu, X.; Che, H.; West, J.S.; Luo, D., 2019. Three Colletotrichum Species, 849 Including a New Species, are Associated to Leaf Anthracnose of Rubber Tree in 850 Hainan, China. Plant Disease 103, 117–124. 851 Corda, A.C.I., 1831. Die Pilze Deutschlands. In: Deutschlands Flora in Abbildungen 852 nach der Natur mit Beschreibungen (Sturm, J, ed.). Sturm, Nürnberg vol. 3, Abt. 853 12: 33–64, tab. 21–32. 854 Costa, J.F.O., Kamei, S.H., Silva, J.R.A., Miranada, A.R.G.S., Netto, M.B., Silva, 855 S.J.C., Correia, K.C., Lima, G.S.A., Assunção, I.P., 2018. Species diversity of 856 Colletotrichum infecting Annona spp. in Brazil. European Journal of Plant 857 Pathology. https://doi.org/10.1007/s10658-018-01630-w. 858 Crouch, J.A., Clarke, B.B., Hillman, B.I, 2009a. What is the value of ITS sequence 859 data in Colletotrichum systematics and species diagnosis? A case study using 860 the falcate-spored graminicolous Colletotrichum group. Mycologia 101, 648–656. 861 Crouch, J.A., Beirn, L.A., Cortese, L.M., Bonos, S.A., Clarke, B.B., 2009b. 862 Anthracnose disease of switchgrass caused by the novel fungal species 863 Colletotrichum navitas. Mycological Research 113, 1411–1421 864 Crouch, J.A., Tredway, L.P., Clarke, B.B. and Hillman, B.I., 2009c. Phylogenetic and 865 population genetic divergence correspond with habitat for the pathogen bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 866 Colletotrichum cereale and allied taxa across diverse grass communities. 867 Molecular Ecology 18, 123–135. 868 Crouch, J.A., Tomaso-Peterson, M., 2012. Anthracnose disease of centipedegrass 869 turf caused by Colletotrichum eremochloae, a new fungal species closely related 870 to Colletotrichum sublineola. Mycologia 104, 1085–1096. 871 Cummings, M.P., Neel, M.C., Shaw, K.L., 2008. A genealogical approach to 872 quantifying lineage divergence. Evolution 62, 2411–2422. 873 Damm, U., Woudenberg, J.H.C., Cannon, P.F., Crous, P.W., 2009. Colletotrichum 874 species with curved conidia from herbaceous hosts. Fungal Diversity 39, 45–87. 875 Damm, U., Cannon, P.F., Liu, F., Barreto, R.W., Guatimosim, E, Crous,P.W., 2013. 876 The Colletotrichum orbiculare species complex: Important pathogens of field 877 crops and weeds. Fungal Diversity 61, 29–59. 878 Damm, U., Cannon, P.F., Woudenberg, J.H.C., Crous, P., 2012a. The Colletotrichum 879 acutatum species complex. Studies in Mycology 73, 37–113. 880 Damm, U., Cannon, P.F., Woudenberg, J.H.C., Johnston, P.R., Weir, B.S., Tan, Y.P., 881 Shivas, R.G., Crous, P.W., 2012b. The Colletotrichum boninense species 882 complex. Studies in Mycology 73:1–36. 883 Damm, U., Sato, T., Alizadeh, A., Groenewald, J.Z., Crous,P.W., 2019. The 884 Colletotrichum dracaenophilum, C. magnum and C. orchidearum species 885 complexes. Studies in Mycology 92, 1–46. 886 Dean, R., Van Kan, J.A.L., Pretorius, Z.A., Hammond-Kosack, K.E., Di Pietro, A., 887 Spanu, P.D., Rudd, J.J., Dickman, M., Kahmann, R., Ellis, J., Foster, G.D., 2012. 888 The Top 10 fungal pathogens in molecular plant pathology. Molecular Plant 889 Pathology 13, 414–430. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 890 Dettman, J.R., Jacobson, D.J., Taylor, J.W., 2003. A multilocus genealogical 891 approach to phylogenetic species recognition in the model eukaryote 892 Neurospora. Evolution 57, 2703–2720. 893 Diao, Y.-Z., Zhang, C., Liu, F., Wang, W.-Z., Liu, L., Cai, L., Liu, X.-L., 2017. 894 Colletotrichum species causing anthracnose disease of chili in China. Persoonia 895 38, 20–37. 896 Doyle, V.P., Oudemans, P.V., Rehner, S.A., Litt, A., 2013. Habitat and host indicate 897 lineage identity in Colletotrichum gloeosporioides s. l. from wild and agricultural 898 landscapes in North America. PLoS ONE 8, e62394. 899 Du, M., Schardl, C., Nuckles, E., Vaillancourt, L., 2005. Using mating-type gene 900 sequences for improved phylogenetic resolution of Collectotrichum species 901 complexes. Mycologia 97, 641–658. 902 Fong, J.J., Fujita, M.K., 2011. Evaluating phylogenetic informativeness and data-type 903 usage for new protein-coding genes across Vertebrata. Molecular Phylogenetics 904 and Evolution 61, 300–307. 905 Fu, M., Crous, P.W., Bai, Q., Zhang, P.F., Xiang, J., Guo, Y.S., Zhao, F.F., Yang, 906 M.M., Hong, N., Xu, W.X. and Wang, G.P., 2019. Colletotrichum species 907 associated with anthracnose of Pyrus spp. in China. Persoonia 42, 1–35. 908 Hebert, P.D., Cywinska, A., Ball, S.L., 2003. Biological identifications through DNA 909 barcodes. Proceedings of the Royal Society of London Series B 270, 313–321. 910 Hyde KD, Cai L, McKenzie EHC, Yang YL, Zhang JZ, Prihastuti, H., 2009. 911 Colletotrichum: a catalogue of confusion. Fungal Diversity 39, 1–17. 912 Hyde, K.D., Udayanga, D., Manamgoda, D.S., Tedersoo, L., Larsson, E., Abarenkov 913 K., Bertrand, Y.J.K., Oxelman, B., Hartmann, M., Kauserud, H., Ryberg, M., 914 Kristiansson, E., Nilsson, R.H., 2013. Incorporating molecular data in fungal bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 915 systematics: A guide for aspiring researchers. Current Research in 916 Environmental and Applied Mycology 3, 1–32. 917 Jayawardena, R.S., Yan, J., Hyde, K.D., Zhang, G.,2016. Morphological and 918 molecular characterization of Colletotrichum species of strawberry in China. 919 Mycosphere 7, 1147–1163. 920 Katoh, K., Misawa, K., Kuma, K., Miyata, T., 2002. MAFFT: a novel method for rapid 921 multiple sequence alignment based on fast Fourier transform. Nucleic Acids 922 Research 30, 3059–3066. 923 Katoh, K., Standley, D.M, 2013. MAFFT multiple sequence alignment software 924 version 7: Improvements in performance and usability. Molecular Biology and 925 Evolution 30, 772–780. 926 Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: Molecular Evolutionary Genetics 927 Analysis version 7.0 for bigger datasets. Molecular Biology and Evolution 33, 928 1870–1874. 929 Larget, B., Kotha, S.K., Dewey, C.N., Ané, C., 2010. BUCKy: Gene tree/species tree 930 reconciliation with the Bayesian concordance analysis. Bioinformatics 26, 2910– 931 2911. 932 Librado, P., Rozas, J.,2009. DnaSP v5: a software for comprehensive analysis of 933 DNA polymorphism data. Bioinformatics 25, 1451–1452. 934 Lima, N.B., Batista, M.V.A., Morais Jr, M.A., Barbosa, M.A.G., Michereff, S.J., Hyde, 935 K.D., Câmara, M.P.S., 2013. Five Colletotrichum species are responsible for 936 mango anthracnose in northeastern Brazil. Fungal Diversity 61:75–88. 937 Liu, F., Cai, L., Crous, P.W., Damm, U., 2014. The Colletotrichum gigasporum 938 species complex. Persoonia 33, 83–97. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 939 Liu, F., Wang, M., Damm, U., Crous, P.W., Cai, L., 2016. Species boundaries in plant 940 pathogenic fungi: A Colletotrichum case study. BMC Evolutionary Biology 16, 81. 941 Liu, F., Weir, B.S., Damm, U., Crous, P.W., Wang, Y., Liu, B., Wang, M., Zhang, M., 942 Cai, L., 2015. Unravelling Colletotrichum species associated with Camellia: 943 Employing ApMat and GS loci to resolve species in the C. gloeosporioides 944 complex. Persoonia 35, 63–86. 945 Lopez-Giraldez, F., Townsend, J.P., 2011. PhyDesign: An online application for 946 profiling phylogenetic informativeness. BMC Evolutionary Biology 11, 152. 947 Magain, N., Miadlikowska, J., Mueller, O., Gajdeczka, M., Truong, C., Salamov, A.A., 948 Dubchak, I., Grigoriev, I.V., Goffinet, B., Sérusiaux, E., Lutzoni, F., 2017. 949 Conserved genomic collinearity as a source of broadly applicable, fast evolving, 950 markers to resolve species complexes: A case study using the lichen-forming 951 genus Peltigera section Polydactylon. Molecular Phylogenetics and Evolution 952 117, 10–29. 953 Marin-Felix, Y., Groenewald, J.Z., Cai, L., Chen, Q., Marincowitz, S., Barnes, I., 954 Bensch, K., Braun, U., Camporesi, E., Damm, U., de Beer, Z.W., Dissanayake, 955 A., Edwards, J., Giraldo, A., Hernandez-Restrepo, M., Hyde, K.D., Jayawardena, 956 R.S., Lombard, L., Luangsa-Ard, J., McTaggart, A.R., Rossman, A.Y., Sandoval- 957 Denis, M., Shen, M., Shivas, R.G., Tan, Y.P., van der Linde, E.J., Wingfield, M.J., 958 Wood, A.R., Zhang, J.Q., Zhang, Y., Crous, P.W., 2017. Genera of 959 phytopathogenic fungi: GOPHY 1. Studies in Mycology 86, 99–216. 960 Mills, P.R., Hodson, A., Brown, A.E., 1992. Molecular differentiation of Colletotrichum 961 gloeosporioides isolates infecting tropical fruits, in: Bailey, J.A., Jeger, M.J. 962 (Eds.), Colletotrichum: Biology, Pathology and Control, CABI, Wallingford, 269– 963 288. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 964 Moriwaki,J., Tsukiboshi,T., 2009. Colletotrichum echinochloae, a new species on 965 Japanese barnyard millet (Echinochloa utilis). Mycoscience 50, 273–280. 966 Niu, X., Gao, H., Qi, J., Chen, M., Tao, A., Xu, J., Dai, Z., Su, J., 2016. Colletotrichum 967 species associated with jute (Corchorus capsularis L.) anthracnose in 968 southeastern China. Scientific Reports 6, 25179. 969 O'Connell, R.J., Thon, M.R., Hacquard, S., Amyotte, S.G., Kleemann, J., Torres, 970 M.F., Damm, U., Buiate, E.A., Epstein, L., Alkan, N., Altmuller, J., Alvarado- 971 Balderrama, L., Bauser, C.A., Becker, C., Birren, B.W., Chen, Z., Choi, J., 972 Crouch, J.A., Duvick, J.P., Farman, M.A., Gan, P., Heiman, D., Henrissat, B., 973 Howard, R.J., Kabbage, M., Koch, C., Kracher, B., Kubo, Y., Law, A.D., Lebrun, 974 M.H., Lee, Y.H., Miyara, I., Moore, N., Neumann, U., Nordstrom, K., Panaccione, 975 D.G., Panstruga, R., Place, M., Proctor, R.H., Prusky, D., Rech, G., Reinhardt, 976 R., Rollins, J.A., Rounsley, S., Schardl, C.L., Schwartz, D.C., Shenoy, N., 977 Shirasu, K., Sikhakolli, U.R., Stuber, K., Sukno, S.A., Sweigard, J.A., Takano, Y., 978 Takahara, H., Trail, F., Van Der Does, H.C., Voll, L.M., Will, I., Young, S., Zeng, 979 Q., Zhang, J., Zhou, S., Dickman, M.B., Schulze-Lefert, P., Van Themaat, E.V., 980 Ma, L.J., Vaillancourt,L.J., 2012. Lifestyle transitions in plant pathogenic 981 Colletotrichum fungi deciphered by genome and transcriptome analyses. Nature 982 Genetics 44, 1060–1065. 983 Oliveira, L.F.M., Feijó, F.M., Mendes, A.L.S.F., Neto, J.D.V., Netto, M.S.B., 984 Assunção, I.P., Lima, G.S.A., 2018. Identification of Colletotrichum species 985 associated with brown spot of cactus prickly pear in Brazil. Tropicala Plant 986 Pathology 43, 247–253. 987 Paradis, E., Claude, J., Strimmer, K., 2004. APE: Analyses of phylogenetics and 988 evolution in R language. Bioinformatics 20, 289–290. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 989 Pond, S.L., Frost, S.D., Muse, S.V., 2005. HyPhy: hypothesis testing using 990 phylogenies. Bioinformatics 21, 676–679. 991 R Core Team. 2017. R: A language and environment for statistical computing. R 992 Foundation for Statistical Computing, Vienna, Austria. 993 Rojas, E.I., Rehner, S.A., Samuels, G.J., Van Bael, S.A., Herre, E.A., Cannon, P., 994 Chen, R., Pang, J., Wang, R., Zhang, Y., Peng, Y.Q., Sha, T., 2010. 995 Colletotrichum gloeosporioides s.l. associated with Theobroma cacao and other 996 plants in Panamá: Multilocus phylogenies distinguish host-associated pathogens 997 from asymptomatic endophytes. Mycologia 102,1318–1338. 998 Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Höhna, S., 999 Larget, B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes v. 3.2: 1000 Efficient Bayesian phylogenetic inference and model choice across a large model 1001 space. Systematic Biology 61, 539–542. 1002 Sakalidis, M.L., Hardy, G.E.S.J., Burgess, T.I., 2011. Use of the Genealogical Sorting 1003 Index (GSI) to delineate species boundaries in the Neofusicoccum parvum- 1004 Neofusicoccum ribis species complex. Molecular Phylogenetics and Evolution, 1005 60, 333–344. 1006 Samarakoon, M.C., Peršoh, D., Hyde, K.D., Bulgakov, T.S., Manawasinghe, I.S., 1007 Jayawardena, R.S., Promputtha, I., 2018. Colletotrichum acidae sp. nov. from 1008 northern Thailand and a new record of C. dematium on Iris sp. Mycosphere 9: 1009 583–597. 1010 Schmitt,.I, Crespo, A., Divakar, P.K., Fankhauser, J.D., Herman-Sackett, E., Kalb, K., 1011 Nelsen, M.P., Nelson, N.A., Rivas-Plata, E., Shimp, A.D.,Widhelm, T., Lumbsch, 1012 H.T., 2009. New primers for promising single-copy genes in fungal phylogenetics 1013 and systematics. Persoonia 23, 35–40. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1014 Sela, I., Ashkenazy, H., Katoh, K., Pupko, T., 2015. GUIDANCE2: Accurate detection 1015 of unreliable alignment regions accounting for the uncertainty of multiple 1016 parameters. Nucleic Acids Research 43, W7–W14. 1017 Sharma, G., Kumar, N., Weir, B.S., Hyde, K.D., Shenoy, B.D., 2013. The ApMat 1018 marker can resolve Colletotrichum species: A case study with Mangifera indica. 1019 Fungal Diversity 61,117–38. 1020 Sharma, G., Pinnaka, A.K., Belle, D.S., 2015. Resolving the Colletotrichum siamense 1021 species complex using ApMat marker. Fungal Diversity 71, 247–64. 1022 Sharma, G., Maymon, M., Freeman, S., 2017. Epidemiology, pathology and 1023 identification of Colletotrichum including a novel species associated with avocado 1024 (Persea americana) anthracnose in Israel. Scientific Reports 7, 15839. 1025 Silva, D.N., Talhinas, P., Várzea, V., Cai, L., Paulo, O.S., Batista, D., 2012. 1026 Application of the Apn2/MAT locus to improve the systematics of the 1027 Colletotrichum gloeosporioides complex: An example from coffee (Coffea spp.) 1028 hosts. Mycologia 104, 396–409. 1029 Silva, J.R.A, Chaves, T.P., Silva,a A.R.G., Barbosa, L.F., Costa, F.F.O., Ramos- 1030 Sobrinho, R., Teixeira, R.R.O., Silva, S.J.C., Lima, G.S.A., Assunção, I.P., 2017. 1031 Molecular and morpho-cultural characterization of Colletotrichum spp. associated 1032 with anthracnose on Capsicum spp. in northeastern Brazil. Tropical Plant 1033 Pathology 42, 315–319. 1034 Sousa, E.S., Silva, J.R.A., Assunção, I.P., Melo, M.P., Feijó, F.M., Matos, K.S., Lima, 1035 G.S.A., Beserra Jr, J.E.A., 2018. Colletotrichum species causing anthracnose on 1036 lima bean in Brazil. Tropical Plant Pathology 43, 78–84. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1037 Sreenivasaprasad, S., Brown, A.E., Mills, P.R., 1992. DNA sequence variation and 1038 interrelationship among Colletotrichum species causing strawberry anthracnose. 1039 Physiological and Molecular Plant Pathology 41, 265–281. 1040 Stawatakis, A., 2014. RAxML version 8: A tool for Phylogenetic Analysis and Post- 1041 Analysis of Large Phylogenies. Bioinformatics 30, 1312–1313. 1042 Sutton, B.C., 1980. The Coelomycetes. Fungi Imperfecti with Pycnidia, Acervuli and 1043 Stromata. CABI, Kew 1044 Tao, G., Liu, Z.Y., Liu, F., Gao, Y.H., Cai,L., 2013. Endophytic Colletotrichum species 1045 from Bletilla ochracea (Orchidaceae), with descriptions of seven new speices. 1046 Fungal Diversity 61, 139–164. 1047 Taylor, J.W., Jacobson, D.J., Kroken, S., Kasuga, T., Geiser, D.M., Hibbett, D.S., 1048 Fisher, M.C., 2000. Phylogenetic species recognition and species concepts in 1049 fungi. Fungal Genetics and Biology 31, 21–32. 1050 Vaidya, G., Lohman, D. J., Meier, R., 2011. SequenceMatrix: concatenation software 1051 for the fast assembly of multi-gene datasets with character set and codon 1052 information. Cladistics 27, 171–180. 1053 Veloso, J.S., Câmara, M.P.S., Lima, W.G., Michereff, S.J., Doyle, V.P.. (2018). Why 1054 species delimitation matters for fungal ecology: Colletotrichum diversity on wild 1055 and cultivated cashew in Brazil. Fungal Biology 122, 677–691. 1056 Vieira, W.A., Michereff, S.J., de Morais Jr, M.A., Hyde, K.D., Câmara, M.P.S., 2014. 1057 Endophytic species of Colletotrichum associated with mango in northeastern 1058 Brazil. Fungal Diversity 67, 181–202. 1059 Vieira, W.A.S., Lima, W.G., Nascimento, E.S., Michereff, S.J., Câmara, M.P.S., 1060 Doyle, V.P., 2017. The impact of phenotypic and molecular data on the inference 1061 of Colletotrichum diversity associated with Musa. Mycologia 109, 912–934. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1062 von Arx, J.A., 1957. Die Arten der Gattung Colletotrichum Cda. Phytopathologische 1063 Zeitschrift 29, 413–468. 1064 Wang, Q.T., Liu, X.T., Ma, H.Y., Shen, X.Y., Hou, C.L., 2019. Colletotrichum 1065 yulongense sp. nov. and C. rhombiforme isolated as endophytes from Vaccinium 1066 dunalianum var. urophyllum in China. Phytotaxa 394, 285–298. 1067 Weir, B.S., Johnston, P.R., Damm, U., 2012. The Colletotrichum gloeosporioides 1068 species complex. Studies in Mycology 73,115–180. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. Table 1. Alignment and phylogenetic informativeness profile statistics for markers used in Colletotrichum by species complex. Table 1. Parsymony 1 1 1 1 -6 2 3 C. acutatum s. l. Length Invariable Variable Singletons PIV (10 ) Pimax informative ACT 239 163 76 24 52 15.36 0.99 CHS-1 251 216 35 5 30 8.56 0.78 GAPDH 272 165 94 18 76 19.95 0.78 HIS3 386 298 88 18 70 24.96 0.67 nrITS 454 427 26 7 19 5.41 0.99 TUB2 481 386 95 24 71 19.7 0.99 C. boninense s. l. ACT 277 199 78 15 63 29.07 0.49 CAL 438 293 142 16 126 52.86 0.58 CHS-1 277 228 49 4 45 22.39 0.48 GAPDH 247 163 83 9 74 47.6 0.24 HIS3 389 292 93 11 82 46.96 0.32 nrITS 537 490 47 11 36 18.41 0.39 TUB2 483 345 137 23 114 48.35 0.63 C. dematium s. l. ACT 235 188 46 1 45 44.65 0.09 CHS-1 251 215 36 6 30 21.48 0.5 GAPDH 264 140 122 13 109 237.56 0.05 HIS3 371 301 70 7 63 183.1 0.05 nrITS 517 492 25 1 24 32.94 0.06 TUB2 497 389 105 10 95 118.42 0.06 C. destructivum s. l. ACT 263 218 44 20 24 18 0.66 CHS-1 280 257 23 11 12 10.13 0.63 GAPDH 225 149 47 16 31 27.35 0.28 HIS3 389 318 71 18 53 38.75 0.37 nrITS 557 526 26 13 13 12.41 0.17 TUB2 514 432 77 33 44 32.49 0.73 C. dracaenophilum s. l. ACT 254 205 49 6 43 31.72 0.44 CHS-1 282 257 25 5 20 18.53 0.46 GAPDH 273 192 81 14 67 52.08 0.41 HIS3 417 354 62 7 55 35.74 0.45 nrITS 541 504 37 13 24 28.58 0.39 TUB2 492 411 81 9 72 58.22 0.4 C. gigasporum s. l. ACT 280 239 37 17 20 23.82 0.19 CAL 675 543 131 56 75 63.14 0.44 CHS-1 299 266 33 15 18 26.38 0.11 GAPDH 290 193 94 51 43 61.02 0.28 GS 702 521 176 92 84 105.17 0.10 HIS3 416 362 53 22 31 45.36 0.12 nrITS 545 500 39 15 24 34.85 0.11 TUB2 537 451 86 40 46 58.50 0.15 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. gloeosporioides s. l. APN2 735 624 108 18 90 73.91 0.17 APN2/MAT-IGS 579 378 201 4 197 149.81 0.19 CAL 648 578 70 1 69 48.76 0.27 GAP2-IGS 708 569 135 14 121 79.71 0.42 GAPDH 767 734 33 2 31 24.51 0.33 GS 665 571 87 24 63 43.83 0.39 TUB2 1231 1116 115 10 105 67.23 0.49 C. graminicola s. l. ACT 269 174 82 52 30 30.25 0.68 CHS-1 280 237 43 24 19 17.06 0.55 nrITS 454 375 70 37 33 30.38 0.23 TUB2 500 350 145 89 56 58.96 0.46 C. magnum s. l. ACT 276 260 16 13 3 11.16 0.40 CHS-1 257 247 10 7 3 11.16 0.40 GAPDH 241 215 26 14 12 13.35 0.42 HIS3 399 360 39 28 11 26 0.44 nrITS 539 533 6 6 0 4.16 0.40 TUB2 536 500 36 24 12 23.15 0.52 C. orbiculare s. l. ACT 226 202 21 5 16 18.37 0.36 CHS-1 280 268 12 2 10 6.16 0.65 GAPDH 243 203 40 8 32 32.01 0.17 GS 954 810 134 41 93 69.13 0.53 HIS3 386 334 50 10 40 36.51 0.13 nrITS 529 514 15 4 11 6.38 0.52 TUB2 467 431 36 9 27 19.75 0.55 C. orchidearum s. l. ACT 276 244 29 17 12 7.06 0.52 CHS-1 265 243 22 6 16 11.46 0.47 GAPDH 242 195 47 37 10 7.99 0.77 HIS3 402 343 59 31 28 16.48 0.67 nrITS 539 521 18 10 8 5.81 0.57 TUB2 540 460 79 51 28 17.44 0.80 C. spaethianum s. l. ACT 210 152 56 28 28 19.44 0.77 GAPDH 240 132 105 28 77 32.53 0.47 nrITS 482 470 10 4 6 8.66 0.73 TUB2 443 351 92 44 48 33.26 0.62 C. truncatum s. l. ACT 215 179 36 6 30 22.89 0.41 GAPDH 276 186 90 6 84 56.86 0.46 nrITS 479 459 20 4 16 10.17 0.53 TUB2 434 353 76 9 67 33.14 0.53 1 - number of base pairs. 2 - PIV, phylogenetic informativeness values on a per net basis 3 - PImax, optimal divergence time. Values are expressed in arbitrary units. 70 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1072 FIGURE CAPTIONS 1073 Fig. 1. Barcode gap and distance overlap between the intra- and inter-specific 1074 distances. Values were calculated based on the intra- and inter-specific distances 1075 frequencies distribution of each Colletotrichum species complex. 1077 Fig. 2. Ultrametric trees and net phylogenetic informativeness profiles of markers 1078 used for phylogenetic studies of 13 Colletotrichum species complexes. Values on the 1079 X-axes correspond to the relative timescale (0—1) based on the root-to-tip distance. -6 1080 Values on the Y-axes represent net phylogenetic informativeness values (10 ) in 1081 arbitrary units. 1083 Fig. 3. Heat map of the Genealogical Sorting Indices (GSI) by Colletotrichum species 1084 complex. GSIs of 1000 bootstrap trees were calculated with 100 permutations. Rows 1085 correspond to species, and columns correspond to individual markers and 1086 concatenated datasets (all markers and best markers). Asterisks represent the best 1087 markers for each complex. 1089 Fig. 4. Primary concordance trees resulting from the Bayesian concordance analyses 1090 including isolates from the C. acutatum complex. A. All markers (ACT, CHS-1, 1091 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1092 Concordance factors are shown above the branches that were resolved by at least 1093 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1095 Fig. 5. Primary concordance trees resulting from the Bayesian concordance analyses 1096 including isolates from C. boninense complex. A. All markers (ACT, CAL, CHS-1, 1097 GAPDH, HIS3, ITS and TUB2). B. Best markers (CAL, GAPDH and TUB2). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1098 Concordance factors are shown above the branches that were resolved by at least 1099 one marker (≥0.14 for all markers and ≥0.33 for the best markers). 1101 Fig. 6. Primary concordance trees resulting from the Bayesian concordance analyses 1102 including isolates from the C. dematium complex. A. All markers (ACT, CHS-1, 1103 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1104 Concordance factors are shown above the branches that were resolved by at least 1105 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1107 Fig. 7. Primary concordance trees resulting from the Bayesian concordance analyses 1108 including isolates from the C. destructivum complex. A. All markers (ACT, CHS-1, 1109 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1110 Concordance factors are shown above the branches that were resolved by at least 1111 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1113 Fig. 8. Primary concordance trees resulting from the Bayesian concordance analyses 1114 including isolates from the C. dracaenophilum complex. A. All markers (ACT, CHS-1, 1115 GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1116 Concordance factors are shown above the branches that were resolved by at least 1117 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1119 Fig. 9. Primary concordance trees resulting from the Bayesian concordance analyses 1120 including isolates from the C. gigasporum complex. A. All markers (ACT, CAL, CHS- 1121 1, GAPDH, GS, HIS3, ITS and TUB2). B. Best markers (CAL, GAPDH and GS). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1122 Concordance factors are shown above the branches that were resolved by at least 1123 one marker (≥0.13 for all markers and ≥0.33 for the best markers). 1125 Fig. 10. Primary concordance trees resulting from the Bayesian concordance 1126 analyses including isolates from the C. gloeosporioides complex. A. All markers 1127 (APN2, APN2/MAT-IGS, CAL, GAPDH, GAP2-IGS, GS and TUB2). B. Best markers 1128 (APN2, APN2/MAT-IGS and GAP2-IGS). Concordance factors are shown above the 1129 branches that were resolved by at least one marker (≥0.14 for all markers and ≥0.33 1130 for the best markers). 1132 Fig. 11. Primary concordance trees resulting from the Bayesian concordance 1133 analyses including isolates from the C. magnum complex. A. All markers (ACT, CHS- 1134 1, GAPDH, HIS3, ITS and TUB2). B. Best markers (GAPDH, HIS3 and TUB2). 1135 Concordance factors are shown above the branches that were resolved by at least 1136 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1138 Fig. 12. Primary concordance trees resulting from the Bayesian concordance 1139 analyses including isolates from the C. orbiculare complex. A. All markers (ACT, 1140 CHS-1, GAPDH, GS, HIS3, ITS and TUB2). B. Best markers (HIS3, GAPDH and 1141 GS). Concordance factors are shown above the branches that were resolved by at 1142 least one marker (≥0.14 for all markers and ≥0.33 for the best markers). 1144 Fig. 13. Primary concordance trees resulting from the Bayesian concordance 1145 analyses including isolates from the C. orchidearum complex. A. All markers (ACT, 1146 CHS-1, GAPDH, HIS3, ITS and TUB2). B. Best markers (CHS-1, HIS3 and TUB2). bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 1147 Concordance factors are shown above the branches that were resolved by at least 1148 one marker (≥0.16 for all markers and ≥0.33 for the best markers). 1150 Fig. 14. Primary concordance trees resulting from the Bayesian concordance 1151 analyses including isolates from the C. spaethianum complex. A. All markers (ACT, 1152 GAPDH, ITS and TUB2). B. Best markers (ACT, GAPDH, and TUB2). Concordance 1153 factors are shown above the branches that were resolved by at least one marker 1154 (≥0.25 for all markers and ≥0.33 for the best markers). 1156 Fig. 15. Primary concordance trees resulting from the Bayesian concordance 1157 analyses including isolates from the C. truncatum complex. A. All markers (ACT, 1158 GAPDH, ITS and TUB2). B. Best markers (ACT, GAPDH, and TUB2). Concordance 1159 factors are shown above the branches that were resolved by at least one marker 1160 (≥0.25 for all markers and ≥0.33 for the best markers). 1163 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. acutatum s. l. C. boninense s. l. C. dematium s. l. C. destructivum s. l. 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 GAPDH GAPDH GAPDH GAPDH CAL ACT ACT HIS3 CHS-1 HIS3 TUB ACT HIS3 TUB2 CHS-1 HIS3 ACT CHS-1 TUB2 CHS-1 TUB2 nrITS nrITS nrITS nrITS 0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.02 0.04 0.06 0.08 0 0,02 0,04 0,06 0,08 0,1 0,12 0 0.05 0.1 0.15 0.2 0.25 0.3 C. dracaenophilum s. l. C. gigasporum s. l. C. gloeosporioides s. l. C. magnum s. l. 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 GS APN2/MAT-IGS GAPDH GAPDH CAL GAP2-IGS ACT HIS3 GAPDH APN2 TUB2 TUB2 TUB2 CAL ACT CHS-1 HIS3 TUB2 CHS-1 ACT CHS-3 GS HIS3 nrITS nrITS GAPDH nrITS 0 0.01 0.02 0.03 0.04 0.05 0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2 0.25 0 0.05 0.1 0.15 C. orchidearum s. l. C. spaethianum s. l. C. truncatum s. l. C. orbiculare s. l. 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100 GS HIS3 GAPDH GAPDH HIS3 CHS-1 GAPDH ACT ACT TUB2 ACT GAPDH TUB2 TUB2 TUB2 ACT CHS-1 nrITS nrITS nrITS nrITS 0 0.005 0.01 0.015 0.02 0.025 0.03 0 0.02 0.04 0.06 0.08 0.1 0 0.05 0.1 0.15 0.2 0.25 0 0.05 0.1 0.15 0.2 0.25 Intra/inter -specific Barcode gap distance overlap bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. simmondsii C. paxtonii C. sloanei C. indonesiense C. lacticiphilum C. brisbanense C. cairnsense C. guajavae C. boninense C. scovillei C. nymphaeae C. torulosum C. walleri C. cybidiicola C. cosmi C. oncidii C. chrysantemi C. beeveri C. melonis C. paranaense C. colombiense C. anthirrhinicola C. limetticola C. brassicicola C. costaricense C. fuscum C. paspali C. annellatum C. tamarilloi C. bryoniicola C. lupini C. phyllanthi C. nicholsonii C. vignae C. cuscutae C. karstii C. graminicola C. abscissum C. higginsianum C. petchii C. navitas C. acutatum C. utrechtense C. fioriniae C. limonicola C. susblineola C. tabacum C. rhobiforme C. novae-zelandiae C. eremochloae C. acerbum C. ocimi C. catinaense C. phormi C. cereale C. destructivum C. parsoniae C. kinghornii C. mischanti C. australe C. americae-borealis C. brasiliense C. salicis C. eleusines C. lini C. hippeastri C. pyricola C. jacksonii C. lentis C. constrictum C. johnstonii C. falcatum C. godetiae C. pisicola C. dacrycarpi 36 55 20 49 18 38 15 33 24 37 13 27 20 30 10 22 16 24 7 16 12 18 5 11 8 12 5 4 0 0 0 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 C. acutatum s. l. C. graminicola s. l. C. boninense s. l. C. destructivum s. l. C. tofieldiae C. verruculosum C. orbiculare C. cliviicola C. liriopes C. sidae C. plurivorum C. chrysophilum C. bletilum C. spinosum C. vittalense C. fructicola C. riograndense C. tebeestii C. orchidearum C. musae C. incanum C. trifolii C. musicola C. siamense C. spaethianum C. malvarum C. cattleyicola C. queenslandicum C. lilii C. bidentis C. soojae C. tropicale C. guizhouensis C. lindemuthianum C. theobromicola C. piperis 45 16 40 14 125 35 12 110 10 94 25 9 78 20 7 62 15 5 47 10 3 31 5 15 0 0 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 C. orchidearum s. l. C. gloeosporioides s. l. C. spaethianum s. l. C. orbiculare s. l. C. lineola C. brevisporum C. arxii C. corchorum-capsularis C. dematium C. lobatum C. vietnamense C. fructi C. truncatum C. merremiae C. radicis C. anthrisci C. panamense C. curcumae C. pseudomajus C. circinans C. magnum C. acidae C. spinaceae C. cacao C. gigasporum C. fusisforme 27 110 24 53 199 47 19 77 16 35 124 29 13 55 10 44 8 33 17 5 22 11 2 11 5 0 0 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 C. dematium s. l. C. magnum s. l. C. gigasporum s. l. C. truncatum s. l. C. yunnanense C. dracaenophilum C. coelogynes C. tropicicola GAP2-IGS nrITS ACT CAL APN2 CHS-1 GS TUB2 GAPDH HIS3 5 APN2/MAT-IGS 0.8 0.6 0.4 0.2 0 C. dracaenophilum s. l. bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. truncatum s. l. C. dracaenophilum s. l. C. magnum s. l. C. gigasporum s. l. C. acidae C. coelogynes C. gigasporum C. brevisporum C. corchorum− C. dracaenophilum C. vietnamense capsularis C. magnum C. curcumae C. tropicicola C. truncatum C. orchidearum s. l. C. orbiculare s. l. C. dematium s. l. C. spaethianum s. l. C. cliviicola C. lindemuthianum C. anthrisci C. guizhoensis C. malvarum C. musicola C. circinans C. lilii C. orbiculare C. orchidearum C. dematium C. liriopes C. plurivorum C. sidae C. lineola C. spaethianum C. spinosum C. sojae C. spinaceae C. tofieldiae C. trifolii C. vittalense C. boninense s. l. C. destructivum s. l. C. gloeosporioides s. l. C. acutatum s. l. C. americae- C. abscissum C. chrysophilum C. boninense borealis C. acutatum C. brasiliense C. destructivum C. fragariae C. australe C. fuscum C. fructicola C. catinaense C. cairnsense C. colombiense C. higginsianum C. musae C. chrysanthemi C. lentis C. constrictum C. queenslandicum C. siamense C. costaricense C. cymbidiicola C. lini C. tabacum C. tropicale C. fioriniae C. hippeastri C. utrechtense C. karstii C. laticiphilum C. vignae C. lupini C. limonicola C. novae−zelandiae C. meleonis C. oncidii C. nymphaeae C. petchii C. paranaense C. torulosum C. paxtonii C. phormii C. scovillei C. simmondsii C. tamarilloi 0 0.25 0.50 0.75 1 GSI values All markers All markers All markers Best markers Best markers Best markers ACT *ACT ACT CHS-1 *CHS-1 *GAPDH *GAPDH nrITS GAPDH *HIS3 *TUB2 *HIS3 nrITS nrITS *TUB2 *TUB2 All markers All markers All markers Best markers Best markers Best markers ACT ACT ACT CHS-1 CHS.1 *CAL *GAPDH CHS-1 *GAPDH *HIS3 *GAPDH *GS nrITS *HIS3 *HIS3 *TUB2 nrITS nrITS TUB2 TUB2 All markers All markers All markers Best markers Best markers Best markers ACT ACT ACT *CAL CHS-1 CHS-1 CHS-1 *GAPDH *GAPDH *GAPDH *HIS3 *GS *HIS3 nrITS HIS3 *TUB2 nrITS nrITS *TUB2 TUB2 All markers All markers All markers Best markers Best markers Best markers ACT *APN2 *ACT *APN2/MAT-IGS CHS-1 *GAPDH CAL *GAPDH GAPDH nrITS *HIS3 *GAP2-IGS *TUB2 nrITS GS *TUB2 TUB2 bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.48 0.73 C. scovillei C. scovillei A 0.37 0.4 B C. guajavae C. guajavae 0.17 0.17 0.58 C. cairnsense C. chrysanthemi 0.21 C. brisbanense 1 0.39 C. cairnsense 0.19 0.33 C. laticiphilum C. cosmi 0.55 0.36 C. indonesiense C. walleri 0.99 0.44 C. cosmi C. nymphaeae Clade 2 0.18 C. walleri 0.74 0.66 0.73 0.21 C. nymphaeae C. paxtonii 0.68 0.73 0.42 0.34 C. simmondsii C. paxtonii Clade 2 0.38 0.37 0.41 C. sloanei 0.17 C. simmondsii 0.33 0.72 C. laticiphilum 0.37 C. sloanei C. indonesiense 0.84 C. chrysanthemi C. brisbanense 0.31 0.73 0.33 C. paranaense C. abscissum 0.33 0.63 0.18 C. limeticola 0.37 0.85 C. costaricense 0.37 C. melonis 0.38 0.35 C. tamarilloi C. costaricense 0.78 0.17 C. lupini 0.74 0.33 C. lupini 0.26 0.18 C. paranaense Clade 1 Clade 1 C. tamarilloi 0.36 1 C. limeticola 0.83 0.38 C. cuscutae C. melonis 0.59 C. abscissum C. cuscutae Clade 3 Clade 3 1 1 C. fioriniae C. fioriniae C. acerbum 1 1 C. acerbum 0.23 C. rhombiforme C. rhombiforme 0.39 0.42 0.44 0.28 C. phormii C. phormii 0.5 0.43 0.49 C. kinghornii C. kinghornii 0.84 0.27 1 0.41 C. australe C. australe Clade 5 Clade 5 C. salicis 0.9 0.95 C. salicis C. johnstonii C. johnstonii 0.69 0.98 0.51 0.67 0.18 C. pyricola C. pyricola 0.42 C. godetiae C. godetiae 0.18 Clade 4 Clade 4 C. acutatum C. acutatum bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.45 0.53 A C. cymbidiicola B C. cymbidiicola 0.34 0.46 0.73 C. oncidii C. oncidii 0.85 1 0.68 0.76 0.4 C. boninense C. boninense 0.56 0.78 0.77 0.85 1 C. torulosum C. torulosum 0.79 1 C. colombiense C. colombiense 0.38 0.46 0.37 0.79 0.77 C. beeveri C. beeveri C. brassicicola 0.21 C. brassicicola 0.88 0.78 C. brasiliense C. constrictum 0.45 1 0.99 C. parsoniae C. dacrycarpi 1 1 C. hippeastri C. brasiliense 0.4 0.52 0.91 C. constrictum C. hippeastri C. dacrycarpi C. parsoniae 0.57 C. catinaense 0.19 0.49 C. karstii 0.69 0.99 C. limonicola 0.26 0.78 C. novae-zelandiae C. phyllanti 0.16 C. annelatum 0.2 C. petchii C. karstii 0.72 0.16 0.5 0.49 C. catinaense 0.36 0.74 0.45 C. phyllanti 0.98 C. limonicola 0.81 C. annelatum 0.38 C. petchii C. novae-zelandiae bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.49 A B 0.62 0.5 0.56 C. dematium C. dematium 0.69 0.35 0.49 0.61 C. lineola C. lineola 0.53 0.58 1 1 C. anthrisci C. anthrisci C. fructi 0.73 0.35 0.56 C. circinans 0.54 1 1 C. circinans 0.49 0.58 C. spinaceae 0.39 C. spinaceae 0.46 C. fructi bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B 0.4 0.35 C. fuscum C. higginsianum 0.34 0.61 0.69 C. vignae C. vignae C. bryoniicola 0.18 C. antirrhinicola 0.29 0.67 0.23 C. fuscum 0.36 0.18 0.35 C. higginsianum 0.41 C. antirrhinicola 0.37 C. bryoniicola 0.27 0.71 C. utrechtense 0.67 C. utrechtense 0.4 0.37 0.34 0.49 0.29 0.39 C. tabacum C. tabacum 0.2 0.99 0.98 C. lentis 0.37 0.41 C. lini 0.39 C. pisicola 0.19 0.28 C. lini 0.53 C. americae 0.28 0.42 0.55 0.83 0.77 C. americae C. destructivum 0.29 C. ocimi 0.99 C. destructivum 0.99 C. lentis 0.37 C. ocimi C. pisicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A 0.41 B C. dracaenophilum 0.88 C. yunnanense C. coelogynes 1 1 C. tropicicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B 0.5 0.73 C. gigasporum C. gigasporum 0.63 0.38 C. pseudomajus C. pseudomajus 1 1 C. vietnamense C. vietnamense 0.37 0.51 0.37 C. radicis C. arxii C. radicis C. arxii bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.52 A B 0.67 C. theobromicola 0.38 C. siamense 0.4 0.65 0.4 1 0.33 C. musae 0.27 0.77 C. queenslandicum 0.26 0.3 0.59 C. siamense 0.56 0.3 C.theobromicola 0.71 1 C. tropicale 0.85 C. queenslandicum 0.27 0.99 0.71 C. tropicale C. chrysophilum 0.78 0.21 0.3 C. chrysophilum C. fructicola 0.88 0.9 0.45 C. musae 0.9 C. fructicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B 0.33 0.46 0.23 0.4 C. brevisporum 0.55 0.67 0.43 0.51 C. lobatum C. merremiae 0.28 0.39 0.72 C. magnum 0.47 0.5 C. panamense C. cacao bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. A B C. sidae 0.33 0.52 C. orbiculare 0.66 0.33 C. tebeestii 0.27 0.79 0.17 C. spinosum 0.24 0.44 C. sidae 0.38 0.36 C. tebeestii 0.39 C. orbiculare 0.44 0.65 C. spinosum 0.59 0.44 0.27 0.34 0.72 1 C. malvarum C. malvarum 0.73 0.88 0.21 0.47 0.39 C. trifolii C. trifolii 0.18 0.17 0.32 0.62 0.62 0.71 C. lindemuthianum C. lindemuthianum 0.32 1 1 0.33 C. bidentis C. bidentis bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. C. cattleyicola A B C. piperis 0.33 0.67 0.71 1 C. sojae 0.5 C. pseudomajus 0.67 1 0.99 C. musicola 0.5 0.67 0.77 1 C. orchidearum 0.74 1 C. vittalense 0.75 1 0.62 0.73 C. plurivorum 0.74 0.81 0.73 0.73 C. cliviicola bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.55 0.71 A B 0.54 0.7 C. spaethianum 0.37 0.5 0.99 0.99 C. lilii 0.75 1 0.5 0.66 0.27 0.36 0.56 0.64 0.75 C. guizhouensis 1 0.94 0.54 C. incanum C. riograndense 0.6 0.73 0.32 0.35 C. tofieldiae 0.99 0.98 0.82 0.77 C. verruculosum 0.99 1 0.58 0.72 0.28 0.35 0.62 C. liriopes 0.77 0.46 C. bletilum bioRxiv preprint first posted online Jun. 4, 2019; doi: http://dx.doi.org/10.1101/659177. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. 0.39 A B 0.33 0.25 C. corchorum- 0.55 capsularis C. truncatum 0.35 0.25 0.33 C. corchorum- 0.4 C. truncatum capsularis 0.96 0.97 0.36 C. truncatum 0.29 C. curcumae C. curcumae 0.83 C. acidae C. acidae 0.98 0.9 C. fusiforme C. fusiforme

Journal

bioRxivbioRxiv

Published: Jun 4, 2019

There are no references for this article.