Meta16S Database

Meta16S: A genus-level 16S rRNA cluster database integrating reference genomes and multi-ecosystem amplicon sequencing data This dataset provides the output of the Meta16S pipeline, a genus-level 16S rRNA cluster database built though incremental clustering of both reference sequences and 16S amplicon sequencing data. Data sources The database integrates sequences from two complementary sources: Reference genomes: 66,360 complete Bacteria and Archaea reference genomes retrieved from NCBI GenBank (one per species, with preference for complete assemblies); 16S rRNA genes were predicted with barrnap, and the V1-V2 hypervariable regions were extracted using seqkit. Amplicon sequencing samples: Human gut microbiome: 691 samples Oral microbiome: 11 samples Environmental water: 32 samples Workflow summary The pipeline is implemented as a Snakemake workflow and proceeds as follow: Quality filtering of raw reads with vsearch (max. expected errors 1.0; length range 100-250 bp; no ambiguous bases allowed). Dereplication of filtered reads per sample. Chimera removal using vsearch. Incremental clustering at 95% sequence identity (genus-level threshold): the reference genome sequences serve as the initial cluster seed; each subsequent sample is mapped to existing cluster centroids, and unmatched sequences are clustered de novo and merged into the growing centroids set. Taxonomy assignment: each cluster is assigned the taxonomy (up to the genus level) of the majority-vote reference genome(s) it contains; clusters with no reference match are labelled Unknown. Output files File Description clusters_map.tsv Two-column table mapping every sequence identifier (reference or amplicon) to its assigned cluster ID. clusters_summary.tsv Per-cluster statistics: cluster ID, number of reference sequences, nummber of amplicon sequences, and assigned taxonomy (up to the genus level). Taxonomic format Taxonomy strings follow the GTDB-style pipe-delimited format: k__<Kingdom>|p__<Phylum>|c__<Class>|o__<Order>|f__<Family>|g__<Genus> Clusters with no reference genome support are assigned Unknown. Source code The pipeline source code is publicly available at https://github.com/cumbof/Meta16S.