def redundant_annotations(go_papers_dict):
go_con, go_cur = gu.open_go()
ancestors_found = {}
to_remove = {}
gpd_leaves_only = {}
for pmid in go_papers_dict.keys(): #[:1000]:
if len(go_papers_dict[pmid]) < 2:
continue
for i in range(len(go_papers_dict[pmid])):
for j in range(len(go_papers_dict[pmid])):
if j >= i: continue
go_id_1 = go_papers_dict[pmid][i]['go_id']
go_id_2 = go_papers_dict[pmid][j]['go_id']
if go_id_1 == go_id_2: continue
sp_id_1 = go_papers_dict[pmid][i]['sp_id']
sp_id_2 = go_papers_dict[pmid][j]['sp_id']
if sp_id_1 != sp_id_2: continue
if gu.is_ancestor(go_id_1, go_id_2, go_cur):
ancestors_found[(pmid,go_id_1,go_id_2)] = \
ancestors_found.get((pmid,go_id_1,go_id_2),0) + 1
to_remove.setdefault(pmid,Set([])).add(j)
#ancestors_found.setdefault(pmid,[]).append(go_id_1,go_id_2)
elif gu.is_ancestor(go_id_2, go_id_1, go_cur):
ancestors_found[(pmid,go_id_2,go_id_1)] = \
ancestors_found.get((pmid,go_id_2,go_id_1),0) + 1
to_remove.setdefault(pmid,Set([])).add(i)
#ancestors_found.setdefault(pmid,[]).append(go_id_2,go_id_1)
go_con.close()
for pmid in go_papers_dict:
if pmid not in to_remove:
gpd_leaves_only[pmid] = go_papers_dict[pmid]
continue
else:
gpd_leaves_only[pmid] = []
for i in range(len(go_papers_dict[pmid])):
if i not in to_remove[pmid]:
gpd_leaves_only[pmid].append(go_papers_dict[pmid][i])
return ancestors_found, to_remove,gpd_leaves_only
def redundant_annotations(go_papers_dict):
go_con, go_cur = gu.open_go()
ancestors_found = {}
to_remove = {}
gpd_leaves_only = {}
for pmid in go_papers_dict.keys(): #[:1000]:
if len(go_papers_dict[pmid]) < 2:
continue
for i in range(len(go_papers_dict[pmid])):
for j in range(len(go_papers_dict[pmid])):
if j >= i: continue
go_id_1 = go_papers_dict[pmid][i]['go_id']
go_id_2 = go_papers_dict[pmid][j]['go_id']
if go_id_1 == go_id_2: continue
sp_id_1 = go_papers_dict[pmid][i]['sp_id']
sp_id_2 = go_papers_dict[pmid][j]['sp_id']
if sp_id_1 != sp_id_2: continue
if gu.is_ancestor(go_id_1, go_id_2, go_cur):
ancestors_found[(pmid,go_id_1,go_id_2)] = \
ancestors_found.get((pmid,go_id_1,go_id_2),0) + 1
to_remove.setdefault(pmid, Set([])).add(j)
#ancestors_found.setdefault(pmid,[]).append(go_id_1,go_id_2)
elif gu.is_ancestor(go_id_2, go_id_1, go_cur):
ancestors_found[(pmid,go_id_2,go_id_1)] = \
ancestors_found.get((pmid,go_id_2,go_id_1),0) + 1
to_remove.setdefault(pmid, Set([])).add(i)
#ancestors_found.setdefault(pmid,[]).append(go_id_2,go_id_1)
go_con.close()
for pmid in go_papers_dict:
if pmid not in to_remove:
gpd_leaves_only[pmid] = go_papers_dict[pmid]
continue
else:
gpd_leaves_only[pmid] = []
for i in range(len(go_papers_dict[pmid])):
if i not in to_remove[pmid]:
gpd_leaves_only[pmid].append(go_papers_dict[pmid][i])
return ancestors_found, to_remove, gpd_leaves_only
#!/usr/bin/env python
import GO.go_utils as gu
import getopt
import sys
opts, args = getopt.getopt(sys.argv[1:],'i:f:o:',
['infile=','field=','outfile='])
gocon, goc = gu.open_go(user="******", passwd="mingus", db="MyGO")
infile = None
go_acc_field = 3 # starts at 1, not 0!
outfile = sys.stdout
for o, a in opts:
if o in ('-i','--infile'):
infile = a
elif o in ('-o','--outfile'):
outfile = open(a,"w")
elif o in ('-f','--field'):
go_acc_field = int(a)
print "infile", infile
if infile:
for inline in file(infile):
sp_id = inline.strip().split()[3]
go_acc = inline.strip().split()[go_acc_field-1]
go_level = gu.go_level(go_acc,goc)
try:
go_term_type = gu.go_acc_to_term_type(go_acc, goc)
except IndexError:
outfile.write("%s\t%s\t%.1f\t%s\n" % (go_acc, "NOTERMTYPE",
go_level, sp_id))
continue
