A multigene phylogeny toward a new phylogenetic classification of Leotiomycetes

Fungi in the class Leotiomycetesare ecologically diverse, including mycorrhizas, endophytes of roots and leaves, plant pathogens, aquatic and aero-aquatic hyphomycetes, mammalian pathogens, and saprobes. These fungi are commonly detected in cultures from diseased tissue and from environmental DNAextracts. The identification of specimens from such character-poor samples increasingly relies on DNA sequencing. However, the current classification of Leotiomycetesis still largely based on morphologically defined taxa, especially at higher taxonomic levels. Consequently, the formal Leotiomycetesclassification is frequently poorly congruent with the relationships suggested by DNA sequencing studies. Previous class-wide phylogenies of Leotiomyceteshave been based on ribosomal DNA markers, with most of the published multi-gene studies being focussed on particular genera or families. In this paper we collate data available from specimens representing both sexual and asexual morphs from across the genetic breadth of the class, with a focus on generic type species, to present a phylogeny based on up to 15 concatenated genes across 279 specimens. Included in the dataset are genes that were extracted from 72 of the genomes available for the class, including 10 new genomes released with this study. To test the statistical support for the deepest branches in the phylogeny, an additional phylogeny based on 3156 genes from 51 selected genomes is also presented. To fill some of the taxonomic gaps in the 15-gene phylogeny, we further present an ITS gene tree, particularly targeting ex- type specimensof generic type species. A small number of novel taxa are proposed: Marthamycetales ord. nov., and Drepanopezizaceae and Mniaeciaceae fams. nov. The formal taxonomic changes are limited in part because of the ad hoc nature of taxon and specimen selection, based purely on the availability of data. The phylogeny constitutes a framework for enabling future taxonomically targeted studies using deliberate specimen selection. Such studies will ideally include designation of epitypes for the type speciesof those genera for which DNA is not able to be extracted from the original type specimen, and consideration of morphological characters whenever genetically defined clades are recognized as formal taxa within a classification.