Abstract
Historically, fungal multigene phylogenies have been reconstructed based on a small number of commonly used genes. The availability of complete fungal genomes has given rise to a new wave of model organisms that provide large number of genes potentially useful for building robust gene genealogies. Unfortunately, cross-utilization of these resources to study phylogenetic relationships in the vast majority of non-model fungi (i.e. “orphan” species) remains an unexamined question. To address this problem, we developed a method coupled with a program named “PHYLORPH” (PHYLogenetic markers for ORPHans). The method screens fungal genomic databases (107 fungal genomes fully sequenced) for single copy genes that might be easily transferable and well suited for studies at low taxonomic levels (for example, in species complexes) in non-model fungal species. To maximize the chance to target genes with informative regions, PHYLORPH displays a graphical evaluation system based on the estimation of nucleotide divergence relative to substitution type. The usefulness of this approach was tested by developing markers in four non-model groups of fungal pathogens. For each pathogen considered, 7 to 40% of the 10–15 best candidate genes proposed by PHYLORPH yielded sequencing success. Levels of polymorphism of these genes were compared with those obtained for some genes traditionally used to build fungal phylogenies (e.g. nuclear rDNA, β-tubulin, γ-actin, Elongation factor EF-1α). These genes were ranked among the best-performing ones and resolved accurately taxa relationships in each of the four non-model groups of fungi considered. We envision that PHYLORPH will constitute a useful tool for obtaining new and accurate phylogenetic markers to resolve relationships between closely related non-model fungal species.
Highlights
Accurate reconstruction of timing and order of species filiations critically depends on obtaining suitable characters for phylogenetic analyses [1]
We introduce a bioinformatic approach to automatically screen genomic databases for single copy protein-coding homologs (SCPCH) that might be well suited for phylogenetic studies, at lower taxonomic level in non-model fungal species
Computational testing The first step in the PHYLORPH method is the identification of single copy protein-coding homologs (SCPCH) by performing a blast search with a protein dataset retrieved from FUNYBASE, PHYLOME-T60 or OrthoMCL-DB against several fungal genomes, close to the non-model taxa under investigation
Summary
Accurate reconstruction of timing and order of species filiations critically depends on obtaining suitable characters for phylogenetic analyses [1]. Transition from single to multi-locus phylogenies has successfully provided novel insights into this field. Topological congruence between multigene phylogenies is commonly used to accurately define phylogenetic species in fungi [2]. Multigene datasets are useful to address questions in population history, demography and speciation. Different genes can evolve in radically different ways resulting in gene/species trees inconsistencies. Using few genes which contain accurate information can allow the construction of robust species trees [1,3,4]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
