The challenge of accurately characterizing metagenome-assembled genomes has been addressed with the introduction of MetaSBT, a novel tool designed for organizing, indexing, and characterizing microbial reference genomes, particularly focusing on viruses. This framework utilizes the Sequence Bloom Tree (SBT) data structure, which employs Bloom filters to efficiently index vast genomic datasets based on their k-mer composition.
In this study, researchers constructed an initial database comprising over 190,000 viral genomes sourced from public repositories. These genomes were systematically grouped into sequence-consistent clusters across seven taxonomic levels, resulting in the identification of over 40,000 candidate species. Notably, approximately 80% of these candidate species do not correspond to any known viral species in existing reference databases.
Framework and Methodology
MetaSBT operates by leveraging a hierarchical framework that organizes microbial genomes into taxonomically consistent clusters. Traditional methods for clustering genomes often rely on phylogenetic analysis, which can be computationally intensive and less effective for large datasets. In contrast, MetaSBT employs a k-mer-based approach that allows for rapid containment queries, significantly reducing the computational burden.
The framework consists of three operational modules: index, profile, and update. The index module retrieves reference genomes and taxonomic metadata, performing quality assessments and constructing a core index from the species level to the kingdom level. The profile module queries uncharacterized genomes against this index, providing a full taxonomic classification based on proximity to known clusters. The update module facilitates the expansion of the database by integrating new genomes without the need for a complete reconstruction.
Results and Implications
The initial database was constructed using 26,285 viral reference genomes from NCBI GenBank, organized into various taxonomic classifications. An additional set of 1,111 viral metagenome-assembled genomes (vMAGs) was incorporated, resulting in the identification of 28 unknown orders, 79 families, 268 genera, and 936 species. Furthermore, the integration of viral sequences from the Metagenomic Gut Virus (MGV) catalog expanded the database significantly, adding 2,976 additional classes, 6,460 orders, 8,622 families, 12,634 genera, and 31,624 species.
MetaSBT’s open-source framework is fully integrated into the Galaxy platform, enhancing accessibility for researchers and facilitating the detection of previously unknown microbial entities. This advancement holds significant promise for the fields of microbial genomics and metagenomics, providing a robust tool for exploring the vast and largely uncharacterized microbial dark matter.
This article was produced by NeonPulse.today using human and AI-assisted editorial processes, based on publicly available information. Content may be edited for clarity and style.
