Python script recovers the genbank ids for all the nucleotide entries linked to a taxon id.

04:59 - Leave a Comment

This python script recovers the genbank ids for all the nucleotide entries linked to a taxon id. The number of requests is minimized using the retmax and retstart parameters provided by the Entrez Utilities.
taxid_2_gbids.py
Python

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

#!/usr/bin/env python


 


import xml.etree.ElementTree as ET


import sys, urllib, urllib2


 


eutils_base_url = "http://eutils.ncbi.nlm.nih.gov/entrez/eutils/"


 


def get_ids(taxid):


    accession_numbers =[]


    retstart = 0


    iteration_step = 10000


    while True:


        result = esearch(db = "nucleotide", term = "txid%s[Organism:exp]"%taxid, retstart = retstart, retmax = iteration_step)


        try:


            result = ET.fromstring(result)


            ids = []


            if result.find('IdList') is not None:


                for id in result.find('IdList').findall('Id'):


                    ids.append(id.text)


 


                result = esummary(db = "nucleotide", ids = ids, retmax = iteration_step)


                result = ET.fromstring(result)


                for docsum in result.findall('DocSum'):


                    for item in docsum.findall("Item[@Name='Caption']"):


                        accession_numbers.append(item.text)


            else:


                break


            


        except Exception, e:


            print e


        retstart += iteration_step


    return accession_numbers


 


def esearch(db, term, retstart = 0, retmax = 20):


    response = urllib.urlopen("%sesearch.fcgi?db=%s&term=%s&retstart=%i&retmax=%i"%(eutils_base_url, db, term, retstart, retmax))


    content = str(response.read())


    response.close()


    return content


 


def esummary(db, ids, retstart = 0, retmax = 20):


    data = {


        'db':db,


        'id':','.join(ids)


    }


    data = urllib.urlencode(data) 


    req = urllib2.Request("%sesummary.fcgi"%eutils_base_url, data)


    response = urllib2.urlopen(req)


    content = str(response.read())


    response.close()


    return content


 


if __name__ == '__main__':


    taxid = None


 


    if "-id" in sys.argv:


        taxid = sys.argv[sys.argv.index("-id")+1]


 


    if not taxid:


        print "Usage: taxid_2_gbids.py -id taxid"


        print "Example: taxid_2_gbids.py -id 4754"


        sys.exit(-1)


 


    ids = get_ids(taxid)


    print ids


    print "%i ids found..."%len(ids)