The genus Pythium (nom. cons.) sensu lato (s.l.) is composed of many important species of plant pathogens. Early molecular phylogenetic studies suggested paraphyly of Pythium, which led to a formal proposal by Uzuhashi and colleagues in 2010 to split the genus into Pythium sensu stricto (s.s.), Elongisporangium, Globisporangium, Ovatisporangium (= Phytopythium), and Pilasporangium using morphological characters and phylogenies of the mt cytochrome c oxidase subunit 2 (cox2) and D1–D2 domains of nuc 28S rDNA. Although the split was fairly justified by the delineating morphological characters, there were weaknesses in the molecular analyses, which created reluctance in the scientific community to adopt these new genera for the description of new species. In this study, this issue was addressed using phylogenomics. Whole genomes of 109 strains of Pythium and close relatives were sequenced, assembled, and annotated. These data were combined with 10 genomes sequenced in previous studies. Phylogenomic analyses were performed with 148 single-copy genes represented in at least 90% of the taxa in the data set. The results showed support for the division of Pythium s.l. The status of alternative generic names that have been used for species of Pythium in the past (e.g., Artotrogus, Cystosiphon, Eupythium, Nematosporangium, Rheosporangium, Sphaerosporangium) was investigated. Based on our molecular analyses and review of the Pythium generic concepts, we urge the scientific community to adopt the generic names Pythium, Elongisporangium, Globisporangium, and their concepts as proposed by Uzuhashi and colleagues in 2010 in their work going forward. In order to consolidate the taxonomy of these genera, some of the recently described Pythium spp. are transferred to Elongisporangium and Globisporangium.

Taken from the Nguyen et al. (2022) paper entitled "Whole genome sequencing and phylogenomic analysis show support for the splitting of genus Pythium.":

Draft genome annotation was performed with the funannotate 1.5.2 (https://github.com/nextgenusfs/funannotate) pipeline following the standard procedure outlined in the “Genome assembly only” tutorial (https://funannotate.readthedocs.io/en/latest/tutorials.html). Briefly, assemblies were repeat-masked with the funannotate mask command, followed by ab initio gene prediction with the funannotate predict command. The funannotate predict command first uses diamond 0.9.21 (Buchfink et al. 2021) and exonerate 2.4.0 (Slater and Birney 2005) to map a set of pre-downloaded proteins from UniProtKB/Swiss-Prot database (release Feb 2019) (Bateman 2019) to the input masked genome assembly. It then runs GeneMark-ES 4.33 (Borodovsky and Lomsadze 2011) to generate one set of gene models, followed by training of Augustus 3.3.1 (Stanke et al. 2008) with BUSCO data leading to the generation of a second set of gene models. EVidenceModeler (EVM) 1.1.1 (Haas et al. 2008) is used to combine the sets of gene models from ab initio gene predictions to give a final set of gene models. The funannotate predict command was run with these options: –busco_db eukaryota_odb9 –busco_seed_species toxoplasma –organism other.

The new genera (Globisporangium, Elongisporangium) proposed by Uzuhashi et al. (2010) have not been used in the file names in this dataset. Please consult with Supplementary Table 1 in Nguyen et al. (2022) to see which Pythium species (s.l.) are in Pythium (s.s.), Globisporangium, and Elongisporangium.

The files can be opened with any text editor (e.g. Notepad, BBEdit) or any command line program (e.g. cat) that can read text files.