extract_genes_ABRicate

Small script to use ABRicate output to extract genes from genome assemblies, reverse complement if necessary, and print to a file

Installation

This script needs Python 3 with the Pandas and BioPython libraries, as well as seqtk to run. ABRicate is not necessarily needed, although the ABRicate output should include a STRAND column with relevent information.

If you have Miniconda installed (https://docs.conda.io/en/latest/miniconda.html), these dependencies can be easily installed. First clone the directory to your machine:

# Clone and enter the directory
git clone https://github.com/boasvdp/extract_genes_ABRicate.git
cd extract_genes_ABRicate

# Create a conda environment with the necessary packages
conda env create -f env.yaml

# Activate the conda environment
conda activate env_extract_genes_ABRicate

Alternatively, these commands can be used to install the tools separately through conda (not in a separate environment!):

conda install -c conda-forge -c bioconda biopython pandas seqtk

Usage

usage: extract_genes_abricate.py [-h] -a ABRICATE FILE -g GENOMES DIR -o OUTPUT DIR [-s SUFFIX] [--genecluster] [--csv] [--flanking]
                                 [--flanking-bp FLANKING LENGTH] [-v]

Extract genes from genes based on ABRicate output.

optional arguments:
  -h, --help            show this help message and exit
  -a ABRICATE FILE, --abricatefile ABRICATE FILE
                        ABRicate file to parse genes
  -g GENOMES DIR, --genomedir GENOMES DIR
                        directory containing genomes
  -o OUTPUT DIR, --output OUTPUT DIR
                        directory for output
  -s SUFFIX, --suffix SUFFIX
                        Genome assembly file suffix (default: .fasta)
  --genecluster         Extract all genes to a single fasta if located on a single contig (default: false)
  --csv                 Use this option if your ABRicate output file is comma-separated (default: parse as tab-separated file).
  --flanking            Extract flanking sequences
  --flanking-bp FLANKING LENGTH
                        Length of flanking sequence to extract in bp (default: 100)
  -v, --verbose         Increase verbosity

IMPORTANT ASSUMPTIONS

The script assumes the genome assemblies are named almost exactly as they are provided in the ABRicate output (#FILE column). The only thing that may differ is the suffix (default .fasta, unless otherwise provided using --suffix). The script is also at this time only able to handle a single suffix for genome assemblies at a time.

If you have identified genes for all genomes in your genomes/ directory (in which all genome assembly files end with .fasta) and your ABRicate output is present in ABRicate_out/strainA.tsv, run:

python extract_genes_ABRicate.py --abricatefile ABRicate_out/strainA.tsv --genomedir genomes/ --output extracted_genes/

Extended usage

ABRicate files can also be combined to speed up things. To combine all files in ABRicate_out/, e.g. run:

cat <(head -n 1 ABRicate_out/strainA.tsv) <(for i in ABRicate_out/*.tsv; do tail -n +2 $i; done) > ABRicate_all.tsv

After which the extract_genes_ABRicate.py script has to be run only once:

python extract_genes_ABRicate.py --abricatefile ABRicate_all.tsv --genomedir genomes/ --output extracted_genes/