fetch-gene-info.py

#!/usr/bin/env python3

"""
Command line wrapper to mygene for bulk operations

Type:

    bulkgene -h

For instructions

See any mygene page eg
http://mygene.info/v3/gene/1017

for list of fields that can be mapped


Example (from symbols):

bulkgene.py -m ensembl.gene -m HGNC -m Vega -m MIM -m alias -m uniprot.Swiss-Prot   ~/tmp/geneset.txt 

From uniprot IDs:

bulkgene.py -s uniprot.Swiss-Prot -s uniprot.TrEMBL -m entrezgene -m ensembl.gene -m HGNC -m Vega -m MIM -m alias -m uniprot.Swiss-Prot /tmp/up

"""

import mygene
import csv
import click
import logging
logging.basicConfig(level=logging.INFO)

@click.command()
@click.option("-t", "--taxon", default="human")
@click.option("-s", "--scopes", multiple=True, default=["symbol"])
@click.option("-m", "--fields", multiple=True, default=["ensembl.gene"])
@click.argument("qfile", type=click.Path(exists=True))
def run(taxon, scopes, fields, qfile):
    qterms = load_input(qfile)
    logging.info('Qterms = {}'.format(len(qterms)))
    mg = mygene.MyGeneInfo()
    payload = mg.querymany(qterms,
                           species=taxon,
                           scopes=scopes,
                           fields=fields,
                           returnall=True)
    rows = payload['out']
    for r in rows:
        vals = [lookup(r, f) for f in fields]
        print("\t".join([ensure_str(v) for v in vals]))

def ensure_str(v):
    if isinstance(v,str):
        return v
    elif isinstance(v,list):
        return "|".join([ensure_str(e) for e in v])
    else:
        return str(v)
    
        
def lookup(obj, path):
    toks = path.split(".")
    v = obj
    for t in toks:
        if t in v:
            v = v[t]
        else:
            return ""
    return v

    
def load_input(qfile):
    syms = []
    with open(qfile, newline='') as csvfile:
        reader = csv.reader(csvfile, delimiter='\t')
        for row in reader:
            t = row[0]
            syms.append(t)
    return syms


if __name__ == "__main__":
    run()