Vinson P. Doyle

Can We Identify Genes with Increased Phylogenetic Reliability

Topological heterogeneity among gene trees is widely observed in phylogenomic analyses and some of this variation is likely caused by systematic error in gene tree estimation. Systematic error can be mitigated by improving models of sequence evolution to account for all evolutionary processes relevant to each gene or identifying those genes whose evolution best conforms to existing models. However, the best method for identifying such genes is not well established. Here, we ask if filtering genes according to their clock-likeness or posterior predictive effect size (PPES, an inference-based measure of model violation) improves phylogenetic reliability and congruence. We compared these approaches to each other, and to the common practice of filtering based on rate of evolution, using two different metrics. First, we compared gene-tree topologies to accepted reference topologies. Second, we examined topological similarity among gene trees in filtered sets. Our results suggest that filtering genes based on clock-likeness and PPES can yield a collection of genes with more reliable phylogenetic signal. For the two exemplar data sets we explored, from yeast and amniotes, clock-likeness and PPES outperformed rate-based filtering in both congruence and reliability.

Untangling the influences of unmodeled evolutionary processes on phylogenetic signal in a forensically important HIV-1 transmission cluster

Stochastic models of sequence evolution have been developed to reflect many biologically important processes, allowing for accurate phylogenetic reconstruction when an appropriate model is selected. However, commonly used models do not incorporate several potentially important biological processes. Spurious phylogenetic inference may result if these processes play an important role in the evolution of a dataset yet are not incorporated into assumed models. Few studies have attempted to assess the relative importance of multiple processes in producing spurious inferences. The application of phylogenetic methods to infer the source of HIV-1 transmission clusters depends upon accurate phylogenetic results, yet there are several relevant unmodeled biological processes (e.g., recombination and convergence) that may cause complications. Here, through analyses of HIV-1 env sequences from a small, forensically important transmission cluster, we tease apart the impact of these processes and present evidence suggesting that convergent evolution and high rates of insertions and deletions (causing alignment uncertainty) led to spurious phylogenetic signal with forensic relevance. Previous analyses show paraphyly of HIV-1 lineages sampled from an individual who, based on non-phylogenetic evidence, had never acted as a source of infection for others in this transmission cluster. If true, this pattern calls into question assumptions underlying phylogenetic approaches to source and recipient identification. By systematically assessing the contribution of different unmodeled processes, we demonstrate that removal of sites likely influenced by strong positive selection both reduces the alignment-wide signal supporting paraphyly of viruses sampled from this individual and eliminates support for the effects of recombination. Additionally, the removal of ambiguously aligned sites alters strongly supported relationships among viruses sampled from different individuals. These observations highlight the need to jointly consider multiple unmodeled evolutionary processes and motivate a phylogenomic perspective when inferring viral transmission histories.

The impact of phenotypic and molecular data on the inference of Colletotrichum diversity associated with Musa

ABSTRACT Developing a comprehensive and reliable taxonomy for the Colletotrichum gloeosporioides species complex will require adopting data standards on the basis of an understanding of how methodological choices impact morphological evaluations and phylogenetic inference. We explored the impact of methodological choices in a morphological and molecular evaluation of Colletotrichum species associated with banana in Brazil. The choice of alignment filtering algorithm has a significant impact on topological inference and the retention of phylogenetically informative sites. Similarly, the choice of phylogenetic marker affects the delimitation of species boundaries, particularly if low phylogenetic signal is confounded with strong discordance, and inference of the species tree from multiple-gene trees. According to both phylogenetic informativeness profiling and Bayesian concordance analyses, the most informative loci are DNA lyase (APN2), intergenic spacer (IGS) between DNA lyase and the mating-type locus MAT1-2-1 (APN2/MAT-IGS), calmodulin (CAL), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), glutamine synthetase (GS), β-tubulin (TUB2), and a new marker, the intergenic spacer between GAPDH and an hypothetical protein (GAP2-IGS). Cornmeal agar minimizes the variance in conidial dimensions compared with potato dextrose agar and synthetic nutrient-poor agar, such that species are more readily distinguishable based on phenotypic differences. We apply these insights to investigate the diversity of Colletotrichum species associated with banana anthracnose in Brazil and report C. musae, C. tropicale, C. theobromicola, and C. siamense in association with banana anthracnose. One lineage did not cluster with any previously described species and is described here as C. chrysophilum.

First report of Anthracnose caused by Colletotrichum sichuanensis on Phaseolus lunatus in Brazil

The lima bean (Phaseolus lunatus) is of cultural importance in the Northeast region of Brazil as source of income and food for the population. Among the diseases impacting lima bean production, anthracnose caused by the genus Colletotrichum is among the most prominent. In 2015, lima-bean leaves with typical symptoms of anthracnose were collected in five states of northeastern Brazil, including Pernambuco, Paraiba, Piaui, Ceara and Alagoas. The symptoms are characterized by reddish spots on the leaf blade and along the primary and secondary veins, resulting in wrinkled leaves. On pods and stems, the lesions are depressed and reddish with developmental acervuli. Small pieces of leaf tissue from the periphery of the lesions were surface disinfested (1 min in 70% alcohol, 2 min in 1% NaOCl, and rinsed with sterile water), plated on potato dextrose agar (PDA), and incubated at 25°C for 7 days. Colonies that were morphologically similar to Colletotrichum spp. were transferred to PDA and preserved. In order to i...

Thiophanate-Methyl Resistance and Fitness Components of Colletotrichum musae Isolates from Banana in Brazil

Anthracnose, caused by Colletotrichum musae, is the most important postharvest disease of banana and is widely distributed among the banana production regions in Brazil. Although thiophanate-methyl is a fungicide frequently used in Brazilian banana orchards to control Sigatoka leaf spot, Collettotrichum populations are also exposed, resulting in the evolution of fungicide resistance and the inability to manage banana anthracnose. We investigated 139 Brazilian isolates of C. musae for thiophanate-methyl sensitivity in vitro. The 50% mycelial growth inhibition (EC50) values varied between 0.003 and 48.73 μg/ml. One-hundred and thirty isolates were classified as sensitive, with EC50 values ranging from 0.003 to 4.84 μg/ml, while the remaining nine isolates were considered moderately resistant, with EC50 values ranging between 10.43 and 48.73 μg/ml. Resistant or highly resistant isolates (EC50 > 100 μg/ml) were not found. A substitution of TAC for TTC at codon 200 in a coding region of the β-tubulin gene was associated with the moderately resistant phenotype. Applications of thiophanate-methyl formulation to detached banana fruit at the label rate (500 μg/ml) showed low efficacy in controlling the moderately resistant isolates on banana fruits. However, there is no indication of a reduction in fitness associated with fungicide resistance as sensitive and moderately resistant isolates do not differ with respect to mycelial growth rate (P = 0.098), spore production (P = 0.066), spore germination (P = 0.366), osmotic sensitivity (P = 0.051), and virulence (P = 0.057). Our results revealed absence of adaptability cost for the moderately resistant isolates, suggesting that they can be dominant in population if the fungicide continue to be applied.

Why species delimitation matters for fungal ecology: Colletotrichum diversity on wild and cultivated cashew in Brazil

Anthracnose is one of the most important plant diseases globally, occurring on a wide range of cultivated and wild host species. This study aimed to identify the Colletotrichum species associated with cashew anthracnose in Brazil, determine their phylogenetic relationships and geographical distribution, and provide some insight into the factors that may be influencing community composition. Colletotrichum isolates collected from symptomatic leaves, stems, inflorescences, and fruit of cultivated and wild cashew, across four Brazilian biomes, were identified as Colletotrichum chrysophilum, Colletotrichum fragariae, Colletotrichum fructicola, Colletotrichum gloeosporioides sensu stricto, Colletotrichum queenslandicum, Colletotrichum siamense and Colletotrichum tropicale. Colletotrichum siamense was the most dominant species. The greatest species richness was associated with cultivated cashew; leaves harbored more species than the other organs; the Atlantic Forest encompassed more species than the other biomes; and Pernambuco was the most species-rich location. However, accounting for the relative abundance of Colletotrichum species and differences in sample size across strata, the interpretation of which community is most diverse depends on how species are delimited. The present study provides valuable information about the Colletotrichum/cashew pathosystem, sheds light on the causal agents identification,and highlights the impact that species delimitation can have on ecological studies of fungi.

Multiple hidden processes complicate phylogenomic inference of deep Basidiomycota relationships

Resolving deep divergences in the fungal tree of life remains a challenging task even for analyses of genome-scale phylogenetic datasets. Relationships between Basidiomycota subphyla, the rusts (Pucciniomycotina), smuts (Ustilaginomycotina) and mushroom forming fungi (Agaricomycotina) represent a particularly challenging situation that posed problems to both traditional multigene and genome-scale phylogenetic studies. Here, we address basal Basidiomycota relationships using three different phylogenomic datasets, concatenated and gene tree-based analyses and examine the contribution of several potential sources of uncertainty, including fast-evolving sites, putative long-branch taxa, model violation and missing data. We inferred conflicting results with different datasets and under different models. Fast-evolving sites and oversimplified models of amino acid substitution favored the grouping of smuts with mushroom-forming fungi, often leading to maximal bootstrap support in both concatenation and Astral analyses. The most conserved datasets grouped rusts with mushroom forming fungi, although this relationship proved labile, sensitive to model choice, different data subsets and missing data. Excluding putative long branch taxa, genes with the highest proportions of missing data and/or genes with strong signal failed to reveal a consistent trend toward one or the other topology, suggesting that additional sources of conflict are at play too. Our analyses suggest that topologies uniting smuts with mushroom forming fungi can arise as a result of inappropriate modeling of amino acid sites that might be prone to systematic bias. While concatenated analyses yielded strong but conflicting support, individual gene trees mostly provided poor support for rusts, smuts and mushroom-forming fungi, suggesting that the true Basidiomycota tree might be in a part of the tree space that is difficult to access using both concatenation and gene tree based approaches. Thus, basal Basidiomycota relationships remain unresolved and might represent a phylogenetic problem that remains contentious even in the genomic era.