OpenPGx is software with useful pharmacogenomics utilities.
It now implements to convert human's genotype to phenotype and annotate with gene-drug recommendations (for all drugs having recommendations in CPIC, DPWG, or FDA pharmacogenomics databases)
$ openpgx <input> [-o <output>]
<input> is a path to JSON file with genotypes to filter recommendations
<output> is a path to where results will be put, in JSON format
Here is an example <input> file that describes person's genotype:
{
"SLCO1B1", "*1A/*1B",
"CYP2D6": "*2≥3/*1≥3",
"HLA-A*31:01": "positive",
"HLA-B*15:02": "negative",
"DPYD": "c.601A>C/c.2194G>A (*6)",
"G6PD": "B (wildtype)"
}
Thank you for using OpenPGx! We really appreciate contrubitions and discussions:
https://github.com/monigenomi/openpgx
import openpgx
openpgx.recommendations({ "SLCO1B1", "*1A/*1B" })Create virtual environment and install dependencies with:
bash setup.sh
Then it's recommended that you use VSCode for development. Tests in CLI can be run with pytest -vv.
If you wish to compute recommendations from raw databases, you can use openpgx update command:
$ openpgx update
Re-computes recommendations database for CPIC, DPWG, and FDA.
You can use this command for example to test and implement new kinds of recommendation.
Options:
--cpic Link or path from which to fetch CPIC recommendations, default:
https://github.com/cpicpgx/cpic-data/releases/download/v1.10/cpic_db_dump-v1.10_inserts.sql.gz
--dpwg Link or path from which to fetch DPWG recommendations, default:
https://api.pharmgkb.org/v1/download/file/data/dosingGuidelines.json.zip
--fda Link or path from which to fetch FDA recommendations, default:
https://raw.githubusercontent.com/PharmGKB/fda-biomarker/master/fda_pgx_associations_table.jsonSome tips:
- Please add tests for each change you make
- Recommendations database generation can be created with
openpgx updatecommand - It's best to use
python -m openpgxduring development to handle module loading properly
EUPL v1.2