gwasCatalogApiDownloader

#! /usr/bin/env python3

# Vivek Rai
# vivekrai@umich.edu
# Parker Lab
#
# Feb 22, 2021

import httpx
import sys
import argparse
import pandas as pd

URI = "https://www.ebi.ac.uk/gwas/summary-statistics/api/studies/{}/associations"


def fetch_associations(accession_id, output):
    response = httpx.get(URI.format(accession_id))
    if response.status_code != 200:
        print(f"Error: {response.status_code}. Check study id.", file=sys.stderr)
        sys.exit(1)

    associations = response.json()["_embedded"]["associations"]
    print(f"Progress: Found {len(associations)} associations..", file=sys.stderr)

    clean_df = []
    keys_of_interest = [
        "variant_id",
        "chromosome",
        "base_pair_location",
        "effect_allele",
        "other_allele",
        "effect_allele_frequency",
        "odds_ratio",
        "beta",
        "p_value",
        "ci_upper",
        "ci_lower",
        "study_accession",
    ]

    for _, v in associations.items():
        _ = {}
        for key in keys_of_interest:
            _val = v.get(key, "NA")
            if _val is None:
                _val = "NA"
            _[key] = v.get(key, "NA")

        _["trait"] = v.get("trait", "[NA]")[0]

        clean_df.append(_)

    pd.json_normalize(clean_df).to_csv(output, index=False, encoding="utf-8", sep="\t")
    return


if __name__ == "__main__":
    parser = argparse.ArgumentParser(
        prog="gwasCatalogApiDownloader",
        usage="""Download summary statistics from GWAS Catalog (ebi.ac.uk) for given accession id.""",
    )
    parser.add_argument(
        "accession_id", metavar="ID", help="Accession ID (GCSTXXXXXX for example)"
    )
    parser.add_argument(
        "-o", "--output", default=sys.stdout, help="output file (default: stdout)"
    )

    args = parser.parse_args()
    fetch_associations(args.accession_id, args.output)
    sys.exit(0)