Simple wrapper pipeline to process a set of genomes in genbank format through roary to identify orthologues and perform pangenome analysis.
The pipeline has been tested with genbank files produced by antiSMASH 5 where CDSs were
predicted by prodigal. In theory, any set of genbank files should work as long as CDS
files are marked as CDS features and they have a locus_tag qualifier.
The pipeline script is nextflow, and it requires python3 with biopython for parsing the genbank files and roary. It was tested with roary version 1.3.
Processes that require python 3 and biopython are labelled 'py3', and those that require roary alre labelled 'roary'. Use the withLabel directive in your nextflow.config file to point those processes to the appropriate installation locations. See an example at the end of the nf file.
After setting your nextflow.config file, there are only three paramters that you
need to specify in the nextflow call:
- indir: Directory with input genbank files. Each genome must correspond to a single
file with extension
.gbk(no compressed files). - outdir: Directory for output.
- roary_threads: Indicate how many threads to use for roary.
Memory/time requrirements can be set in the nextflow.config file.