Bakta database

This data repository contains the mandatory DB for Bakta.

It is available in two versions: the default (db.tar.gz or) and a lightweight alternative (db-light.tar.gz).

Bakta is a tool for the rapid & standardized local annotation of bacterial genomes & plasmids. It provides dbxref-rich and sORF-including annotations in machine-readble JSON & bioinformatics standard file formats for automatic downstream analysis: https://github.com/oschwengers/bakta

This db provides protein sequence hash digests and lengths of UniProt's UniRef100 clusters, UniParc and NCBI RefSeq sequences for ultra-fast identification & lookups. It has been pre-annotated with several specialized db and enriched with Dbxrefs. Furthermore, seed sequences of UniProt's UniRef90 clusters are stored for fallback homology searches via Diamond sequence alignments. All conducted pre-annotations are logged and provided in the db.log.gz file.

External DB versions:

NCBI AMRFinderPlus: 2024-12-18.1

COG: 2024

DoriC: 12

ISFinder: 2019-09-25

Mob-suite: 3

Pfam: 37.2

RefSeq: r228

Rfam: 15

UniProtKB/Swiss-Prot: 2025_01

VFDB: 2025-02-14