Classifying prokaryotic genomes using the Microbial Genomes Atlas (MiGA) webserver.

Abstract

The 16S rRNA gene has served effectively as the backbone of prokaryotic taxonomy and phylogeny but offers limited resolution at the species level and is often not assembled as part of metagenome‐assembled genomes (MAGs) recovered from environmental DNA or single‐cell amplified genomes (SAGs). The genome‐aggregate average nucleotide identity (ANI) and amino acid identity (AAI) have been shown to provide robust resolution at the species and subspecies levels and can be reliably calculated based on as little as ∼200 kb of a genome subsample. To identify genomes of already described taxa as well as assess the degree of novelty of genomes representing novel taxa and aid their taxonomic descriptions based on the ANI/AAI concept, we have developed the Microbial Genomes Atlas (MiGA) webserver (www.microbial-genomes.org). Therefore, MiGA represents the genome equivalent of ribosomal webservers. MiGA is regularly updated to include all available genomes of cultured and uncultured (Candidatus) type material in its reference database and offers additional bioinformatics services such as genome assembly from unassembled reads, quality assessment, and detection of probable horizontal gene transfer events from close relatives. In this article, we provide practical examples of how to use MiGA and interpret its results. The advantages of MiGA over alternative whole‐genome‐based approaches are also summarized.