ghettoblast

Dependencies

This workflow have been developped with the following versions, previous versions have not been tested

• Snakemake 6.13.1
• Python 3.6.5
• conda 4.6.14

Setup

In order to make the pipeline work, fill following text field, copy it into a plain text file and save it as config/config.yaml

query_file:
        /path/to/target_seqs.fasta
sample_path:
        /path/to/sample_dir
outdir:
        /path/to/output_dir
exclude_seqs:
        FALSE
top_only:
        TRUE

When exclude_seqs is TRUE, query and subject sequences will be extracted from the final blast matrix, and the subject sequences saved as individual fasta files. When top_only is TRUE, the final blast matrix only contains tophits for each subject sequences against each query sequences.

The workflow automatically scans the sample_dir for .fasta files and will attempt to include these in the workflow, using their file name as sample names (excluding the .fasta extension.

Example:

> ls /path/to/sample_dir/
S_aureus1.fasta  S_aureus2.fasta  C_freundii1.fasta

Their sample name will be S_aureus1 S_aureus2, and C_freundii1

Sample exclusion

If you wish to exclude specific samples within the sample_dir/, provide sample names into the config/blacklist.tsv file. Make sure to maintain sample as the header, and write one sample name per line (note don't use their full file names!). Example: to exclude C_freundii1 the config/blacklist.tsv file should look like this

sample
C_freundii1

Execution

To execute the workflow first activate the snakemake environment, then specify the amounts of cores to use (for parallization) with --cores X and add the --use-conda so that the correct environments can be generated and activated.

> cd ghettoblast
> conda activate snakemake
> snakemake --cores 3 --use-conda