def go_terms_with_ec_per_paper(papers,outpath=None,top=20):
"""Create a dict that counts up how many times a specific (GO ID, GO Term Text, EvCode)
tuple occurs for each paper"""
#
# Can be used with SP & GOA data
go_ec_count = {}
go_con = mysqlConnect()
go_cur = go_con.cursor()
for p in papers:
for rec in papers[p]:
go_id = rec['go_id']
go_ec = rec['go_ec']
try:
name = gu.go_acc_to_name(go_id,go_cur)
except IndexError: #sometimes the GO ID given is actually a synonym
try:
name = gu.go_acc_to_synonym_name(go_id, go_cur)
except IndexError: #sometimes it just doesn't work
print "problem with GO ID", go_id
name = ''
gokey = (go_id, name, go_ec)
# go_ec_count[PMID] = {{(GO ID, GO Term Text, Ev Code) : # times paper gives this annotaion}}
if p in go_ec_count:
go_ec_count[p][gokey] = go_ec_count[p].get(gokey,0) + 1
else:
go_ec_count[p] = {gokey: 1}
go_con.close()
return go_ec_count
def go_terms_per_paper(papers,outpath=None,top=20):
#
# Can be used with SP & GOA data
go_count = {}
go_con = mysqlConnect()
go_cur = go_con.cursor()
for p in papers:
for rec in papers[p]:
go_id = rec['go_id']
try:
name = gu.go_acc_to_name(go_id,go_cur)
except IndexError: #sometimes the GO ID given is actually a synonym
try:
name = gu.go_acc_to_synonym_name(go_id, go_cur)
except IndexError: #sometimes it just doesn't work
print "problem with GO ID", go_id
name = ''
gokey = (go_id, name)
# go_count[PMID] = {{(GO ID, GO Term Text) : # times paper gives this annotaion}}
if p in go_count:
go_count[p][gokey] = go_count[p].get(gokey,0) + 1
else:
go_count[p] = {gokey: 1}
go_con.close()
return go_count
def go_terms_with_ec_per_paper(papers, outpath=None, top=20):
"""Create a dict that counts up how many times a specific (GO ID, GO Term Text, EvCode)
tuple occurs for each paper"""
#
# Can be used with SP & GOA data
go_ec_count = {}
go_con = mysqlConnect()
go_cur = go_con.cursor()
for p in papers:
for rec in papers[p]:
go_id = rec['go_id']
go_ec = rec['go_ec']
try:
name = gu.go_acc_to_name(go_id, go_cur)
except IndexError: #sometimes the GO ID given is actually a synonym
try:
name = gu.go_acc_to_synonym_name(go_id, go_cur)
except IndexError: #sometimes it just doesn't work
print "problem with GO ID", go_id
name = ''
gokey = (go_id, name, go_ec)
# go_ec_count[PMID] = {{(GO ID, GO Term Text, Ev Code) : # times paper gives this annotaion}}
if p in go_ec_count:
go_ec_count[p][gokey] = go_ec_count[p].get(gokey, 0) + 1
else:
go_ec_count[p] = {gokey: 1}
go_con.close()
return go_ec_count
def go_terms_per_paper(papers, outpath=None, top=20):
#
# Can be used with SP & GOA data
go_count = {}
go_con = mysqlConnect()
go_cur = go_con.cursor()
for p in papers:
for rec in papers[p]:
go_id = rec['go_id']
try:
name = gu.go_acc_to_name(go_id, go_cur)
except IndexError: #sometimes the GO ID given is actually a synonym
try:
name = gu.go_acc_to_synonym_name(go_id, go_cur)
except IndexError: #sometimes it just doesn't work
print "problem with GO ID", go_id
name = ''
gokey = (go_id, name)
# go_count[PMID] = {{(GO ID, GO Term Text) : # times paper gives this annotaion}}
if p in go_count:
go_count[p][gokey] = go_count[p].get(gokey, 0) + 1
else:
go_count[p] = {gokey: 1}
go_con.close()
return go_count
def top_ontology(papers,outpath=None,top=20):
"""Determines the top GO terms annotated in the analysis set and 1) puts it in
the output dict top_go and 2) writes it out to a tab delim file 'outpath'
Note: this function is currently identical to top_go_terms()"""
#
# Can be used with SP & GOA data
go_count = {}
for p in papers:
for rec in papers[p]:
go_id = rec['go_id']
go_count[go_id] = go_count.get(go_id,0) + 1
top_go = [(i[1],i[0]) for i in go_count.items()]
top_go.sort()
if outpath:
go_con = mysqlConnect()
go_cur = go_con.cursor()
f = open(outpath,"w")
for i in top_go[-top:]:
name = gu.go_acc_to_name(i[1],go_cur)
f.write("%d\t%s\t%s\n" % (i[0], i[1], name))
go_hist = {}
for i in top_go:
go_hist[i[0]] = go_hist.get(i[0],0) + 1
go_hist_list = [(h[1],h[0]) for h in go_hist.items()]
go_hist_list.sort()
fhist = open("hist_%s" % outpath, "w")
for h in go_hist_list:
# print h
fhist.write("%d\t%d\n" % h)
f.close()
fhist.close()
go_con.close()
return top_go
def top_ontology(papers, outpath=None, top=20):
"""Determines the top GO terms annotated in the analysis set and 1) puts it in
the output dict top_go and 2) writes it out to a tab delim file 'outpath'
Note: this function is currently identical to top_go_terms()"""
#
# Can be used with SP & GOA data
go_count = {}
for p in papers:
for rec in papers[p]:
go_id = rec['go_id']
go_count[go_id] = go_count.get(go_id, 0) + 1
top_go = [(i[1], i[0]) for i in go_count.items()]
top_go.sort()
if outpath:
go_con = mysqlConnect()
go_cur = go_con.cursor()
f = open(outpath, "w")
for i in top_go[-top:]:
name = gu.go_acc_to_name(i[1], go_cur)
f.write("%d\t%s\t%s\n" % (i[0], i[1], name))
go_hist = {}
for i in top_go:
go_hist[i[0]] = go_hist.get(i[0], 0) + 1
go_hist_list = [(h[1], h[0]) for h in go_hist.items()]
go_hist_list.sort()
fhist = open("hist_%s" % outpath, "w")
for h in go_hist_list:
# print h
fhist.write("%d\t%d\n" % h)
f.close()
fhist.close()
go_con.close()
return top_go
