This Python script compares variants from a VCF file against a list of known mutations provided in a TSV file. It generates an output TSV file containing the original TSV data with additional 'found_vaf' and 'detected' columns, and a plot comparing Expected VAF and Found VAF.
This tool assumes that you have a known_variants.tsv such as:
| chr | position | ref | alt | gene | hgvsp | vaf |
|---|---|---|---|---|---|---|
| chr1 | 115256530 | G | T | NRAS | p.Q61K | 12.5% |
| chr3 | 41266101 | C | A | CTNNB1 | p.S33Y | 32.5% |
| chr3 | 41266133 | CCTT | C | CTNNB1 | p.S45del | 10.0% |
| chr3 | 178936091 | G | A | PIK3CA | p.E545K | 9.0% |
| chr3 | 178952085 | A | G | PIK3CA | p.H1047R | 17.5% |
| chr4 | 55599321 | A | T | KIT | p.D816V | 10.0% |
| chr4 | 55602765 | G | C | KIT | p.L862L | 7.5% |
| chr5 | 112175770 | G | A | APC | p.T1493T | 35.0% |
| chr7 | 55241707 | G | A | EGFR | p.G719S | 24.5% |
And an input.vcf to be tested whether if the variants have beed detected or not.
- Install the required Python packages:
pip3 install pandas seaborn matplotlib numpy
- Clone the repository
git clone https://github.com/GENCARDIO/eval_somatic_control.git
cd eval_somatic_control
To run the script, use the following command:
python eval_somatic_control.py --input_vcf sample.vcf --known_tsv known_variants.tsv --output_tsv output_file.tsv
Arguments
--input_vcf: Input VCF file containing the query variants.
--known_tsv: Input TSV file containing the list of known mutations.
--output_tsv: Output TSV file with additional columns.