ComBact 0.8

ComBact (Compare Bacteria) is a set of python scripts to highlight genetic differences between multiple bacterial strains of the same species (high sequences similarity is required).

ComBact is based on the idea of creating a local database using a set of genomes of interest against which a set of genes is aligned, exploiting the accuracy and speed of the BLAST algorithm. The reference genome must be in FASTA format (otherwise see point (1) of [Usage]).

Mutations nomenclature is loosely based on HGVS raccomendations (http://varnomen.hgvs.org/)

Prerequisites

• NCBI Blast+ >= 2.7.0
• Python2
• Biopython

Installation

1. Clone the package or download the latest release and unpack the .tar.gz archive.

2. Add ComBact folder to your PATH with export PATH=/path/to/combact:$PATH (add the line to your .bashrc file to avoid typing it every time).

Usage

1. If using a reference in Genbank or Ensembl format, first you must extract genes (and, optionally, intergenic regions) using parse_gbk.py. Required options are the reference file, and the output file name;

2. Align your reference to your set of genomes using blastdb_and_blastn.py. First a "blast database" is created to optimize the alignment process. Then the alignment process starts and the results are saved in two files: an ASN.1 blast format, from which users can create their preferred output format using blast_formatter, and an XML output which is then parsed by the third script. Required options are the fasta file with the reference genes, the name of the output file (without any extension) and the text file reporting the absolute paths to each genome to analyze. The latter will be used in the next step too.

3. Run combact.py to produce the three tables reporting the mutation status of each gene in the selected genomes. Required options are the XML file produced by blastdb_and_blastn.py, the output folder where the output tables are created, and the text file listing the genomes to be analyzed. Users can provide subsets of such list to generate tables with only a selection of genomes.

Options

Users can set length (-L) and identity (-I) thresholds to avoid reporting genes/fragments with alignment scores below given thresholds.

Example of mutations

• SNP: c.[64C>T;102T>G]
• INDEL: c.[64_66delG;102T>G]
• FRAGMENT: c.64C>T;102T>G