vClassifier: a toolkit for species-level classification of prokaryotic viruses
Zhou, K.; Kosmopoulos, J. C.; Anantharaman, K.
Show abstract
As the most abundant and diverse biological entities, prokaryotic viruses play pivotal roles in ecological systems. Their taxonomic classification has been instrumental in elucidating their diversity and ecological functions. However, determination of viral taxonomy remains a considerable challenge. Recently developed approaches succeed in assignment of viral taxonomy at higher ranks, such as at the family level and above, but struggle at the subfamily level and below to the genus and species resolutions. Here, we describe the vClassifier toolkit, a phylogeny-informed methodology to provide species-level taxonomic assignments of viruses. We used single-copy marker genes relevant to specific taxa and reference phylogenetic trees for these groups which facilitates direct comparisons with the taxonomic framework of the International Committee on Taxonomy of Viruses (ICTV). Our method demonstrated significant congruence with the ICTV taxonomy, showing 84-91% alignment at the subfamily and genus levels. For species-level classification, our strategy was integrated with average nucleotide identity, yielding a high congruence rate of over 92% with the taxonomic data from the NCBI Virus database. Benchmarking comparisons revealed that vClassifier matches or surpasses other available tools regarding precision and classification success rates. By achieving objectivity and high levels of consistency, vClassifier streamlines the taxonomic categorization of prokaryotic viral genomes. Accurate assignments at the subfamily, genus, and species levels will significantly refine the taxonomic resolution of viruses, fostering a deeper understanding of viral diversity in microbiomes and ecosystems.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- Machine learning using intrinsic genomic signatures for rapid classification of novel pathogens: COVID-19 case study 96%
- Short k-mer Abundance Profiles Yield Robust Machine Learning Features and Accurate Classifiers for RNA Viruses 95%
- The gut virome of healthy children during the first year of life is diverse and dynamic 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.