nf-bakta

A Nextflow pipeline for the annotation of bacterial genomes or MAGs running Bakta.

Introduction

Bakta is a rapid and standardized tool for annotating bacterial genomes or MAGs obtained from metagenomic samples. It generates as output a set of standard bioinformatics files for further analysis (e.g. tsv, gff3, gbff, fna among other format).

Requirements

Running Bakta requires the download and curation of mandatory databases to use them internally (latest v5.0 are full: 60GB and light: 3GB). Both databases comprise sequences and models from UniProt, NCBI RefSeq, AMRFinder and Pfam universe, being the full type more comprehensive in taxonomic annotation.

Pipeline usage/parameters

The input for this pipeline is a set of FASTA files containing the genomic sequences in a formatted csv file (see input examples).

Usage:
  --input  directory of genomes/MAGs files (fasta format, i.e .fna)
           [default: null]

Optional arguments:
  --outdir results directory
           [default: results]
  --db     database type {path-to/db-full|db-light}
           [default: db-light]

Nextflow execution

Examples for metagenomic samples (DiGenomaLab environment):

1. Run the workflow using a small dataset in the local node (singularity).

nextflow run main.nf -c bakta.config --input input.mini.csv -profile singularity

2. Run in background mode with the full dataset of samples, using the full database in distributed nodes (kutral cluster). Consider the resume option for caching the tasks in case of any error.

nextflow run main.nf -c bakta.config -bg --input input.csv --db db/db-full -profile kutral {-resume}

For configuration details, please check/modify the bakta.conf file.

Software versions

• Bakta: 1.8.1
• Nextflow: 23.04.2