def filter_in_experimental(handle):
outhandle = open(handle.name + ".exp_evidence", "w")
outhandle.write('!gaf-version: 2.0\n')
for inrec in upg.gafiterator(handle):
if upg.record_has(inrec, {'Evidence': GO_EXP_EC}):
upg.writerec(inrec, outhandle)
outhandle.close()
def filter_in_IEA(handle):
outhandle = open(handle.name + ".IEA", "w")
outhandle.write('!gaf-version: 2.0\n')
for inrec in upg.gafiterator(handle):
if inrec['Evidence'] == 'IEA':
upg.writerec(inrec, outhandle)
outhandle.close()
def extract_gaf(rec, outfile, GAFFIELDS, record, sp_id, taxon):
t = ()
if sp_id.has_key(rec['DB_Object_ID']):
GOAParser.writerec(rec, outfile, GAFFIELDS)
if len(GAFFIELDS) == 15:
t = (rec['DB'], rec['DB_Object_ID'], rec['DB_Object_Symbol'],
('|'.join(rec['Qualifier'])), rec['GO_ID'],
('|'.join(rec['DB:Reference'])), rec['Evidence'],
('|'.join(rec['With'])), rec['Aspect'], rec['DB_Object_Name'],
('|'.join(rec['Synonym'])), rec['DB_Object_Type'],
('|'.join(rec['Taxon_ID'])), rec['Date'], rec['Assigned_By'])
elif len(GAFFIELDS) == 17:
t = (rec['DB'], rec['DB_Object_ID'], rec['DB_Object_Symbol'],
('|'.join(rec['Qualifier'])), rec['GO_ID'],
('|'.join(rec['DB:Reference'])), rec['Evidence'],
('|'.join(rec['With'])), rec['Aspect'],
('|'.join(rec['DB_Object_Name'])), ('|'.join(rec['Synonym'])),
rec['DB_Object_Type'], ('|'.join(rec['Taxon_ID'])),
rec['Date'], rec['Assigned_By'], rec['Annotation_Extension'],
rec['Gene_Product_Form_ID'])
record.append(t)
return record
def all_exclusive_IEA(handle):
outhandle = open(handle.name + ".exclusive_IEA", "w")
outhandle.write('!gaf-version: 2.0\n')
for protrec in upg.gafbyproteiniterator(handle):
if exclusive_IEA(protrec):
for outrec in protrec:
upg.writerec(outrec, outhandle)
outhandle.close()
def all_hasnt_experimental(handle):
outhandle = open(handle.name + ".noexp", "w")
outhandle.write('!gaf-version: 2.0\n')
for protrec in upg.gafbyproteiniterator(handle):
if not has_experimental(protrec):
for outrec in protrec:
upg.writerec(outrec, outhandle)
outhandle.close()
def test_selection_writing(self):
"""Test record_has, and writerec.
Adapted from Bio.UniProt.GOA.py by Iddo Friedberg [email protected].
"""
recs = []
filtered = []
# Fields to filter
evidence = {"Evidence": {"ND"}}
synonym = {"Synonym": {"YA19A_YEAST", "YAL019W-A"}}
taxon_id = {"Taxon_ID": {"taxon:559292"}}
# Temporal file to test writerec
f_number, f_filtered = tempfile.mkstemp()
os.close(f_number)
# Open a file and select records as per filter
with open("UniProt/goa_yeast.gaf", "r") as handle:
for rec in GOA.gafiterator(handle):
recs.append(rec)
# Filtering
if (
GOA.record_has(rec, taxon_id)
and GOA.record_has(rec, evidence)
and GOA.record_has(rec, synonym)
):
filtered.append(rec)
# Check number of filtered records
self.assertEqual(len(filtered), 3)
# Write the filtered records to a file using writerec
with open(f_filtered, "w") as handle:
# '!gaf-version: 2.1'
handle.write("!gaf-version: 2.1 \n") # Adding file header
for rec in filtered:
GOA.writerec(rec, handle)
# Open and read the file containing the filtered records
recs_ff = [] # Records from filtered file
with open(f_filtered, "r") as handle:
for rec in GOA.gafiterator(handle):
recs_ff.append(rec)
# Delete test file
os.remove(f_filtered)
# Compare, recs saved by writerec and filtered recs
self.assertEqual(filtered, recs_ff)
def extract_taxon(handle, in_taxid):
"""
Create a GAF file from a single taxon
"""
header = "!gaf-version: 2.0\n"
if isinstance(in_taxid, int):
taxid = str(in_taxid)
taxid = in_taxid.strip()
outfile = open("%s.taxon.%s" % (handle.name, taxid), 'w')
outfile.write(header)
for inrec in upg.gafiterator(handle):
if inrec['Taxon_ID'][0].split(':')[1] == taxid:
upg.writerec(inrec, outfile)
outfile.close()
def extract_taxa(handle, taxalist):
"""
Create a GAF file from multiple taxa
taxalist is a list of strings of taxid. Don't use list of int
"""
outfiles = {}
header = "!gaf-version: 2.0\n"
for taxid in taxalist:
outfiles[taxid] = open("%s.taxon.%s" % (handle.name, taxid), 'w')
outfiles[taxid].write(header)
for inrec in upg.gafiterator(handle):
cur_taxid = inrec['Taxon_ID'][0].split(':')[1]
if cur_taxid in taxalist:
upg.writerec(inrec, outfiles[cur_taxid])
for i in outfiles:
outfiles[i].close()
def split_to_ontologies(handle):
"""Splits a GAF file into three ontology files
"""
header = "!gaf-version: 2.0\n"
out_mfo = open("%s.MFO" % handle.name, 'w')
out_bpo = open("%s.BPO" % handle.name, 'w')
out_cco = open("%s.CCO" % handle.name, 'w')
out_bpo.write(header)
out_mfo.write(header)
out_cco.write(header)
for inrec in upg.gafiterator(handle):
if inrec['Aspect'] == 'F':
upg.writerec(inrec, out_mfo)
elif inrec['Aspect'] == 'P':
upg.writerec(inrec, out_bpo)
elif inrec['Aspect'] == 'C':
upg.writerec(inrec, out_cco)
else:
raise ValueError, 'unknown ontology aspect %s' % inrec['Aspect']
out_mfo.close()
out_bpo.close()
out_cco.close()
#!/usr/bin/env python
import sys
import argparse
import target_prep as tp
from Bio.UniProt import GOA as upg
if __name__ == '__main__':
# parser = argparse.ArgumentParser(description='Filter by field')
# parser.add_argument('-o','--output')
# parser.add_argument('-f','--field')
outhandle = sys.stdout
if len(sys.argv) == 5:
outhandle = open(sys.argv[4],"w")
outhandle.write('!gaf-version: 2.0\n')
goodvals = {sys.argv[1]: set(sys.argv[2].split(','))}
for inrec in upg.gafiterator(open(sys.argv[3])):
if upg.record_has(inrec, goodvals):
upg.writerec(inrec,outhandle)