elif opts.slim_out == "all":
only_direct = False
else:
p.print_help()
sys.exit(1)
# load DAGs
go_dag = GODag(obo_file)
goslim_dag = GODag(slim_obo_file)
# in case a single term is given as input:
if opts.term:
if opts.term not in go_dag:
print(("term %s not found!" % opts.term), file=sys.stderr)
sys.exit(1)
direct_anc, all_anc = mapslim(opts.term, go_dag, goslim_dag)
# output either all or only direct slims, depending on user command
if only_direct:
slim_terms_str = ";".join(direct_anc)
else:
slim_terms_str = ";".join(all_anc)
print(slim_terms_str)
# in case a association file is given as input
if opts.ass_file_name:
assert os.path.exists(opts.ass_file_name), ("file %s not found!"
% opts.ass_file_name)
assocs = read_associations(opts.ass_file_name)
for protein_product, go_terms in assocs.items():
all_direct_anc = set()
all_covered_anc = set()
expected_results = {
'GO:0000005': (set(['GO:0000002', 'GO:0000003']),
set(['GO:0000001', 'GO:0000002', 'GO:0000003'])),
'GO:0000006': (set(['GO:0000003']), set(['GO:0000001', 'GO:0000003'])),
'GO:0000007':
(set(['GO:0000004']), set(['GO:0000001', 'GO:0000003', 'GO:0000004'])),
'GO:0000008': (set(['GO:0000003']), set(['GO:0000001', 'GO:0000003'])),
'GO:0000009':
(set(['GO:0000004']), set(['GO:0000001', 'GO:0000003', 'GO:0000004'])),
'GO:0000010': (set(['GO:0000002', 'GO:0000003']),
set(['GO:0000001', 'GO:0000002', 'GO:0000003']))
}
tests_succeed = True
for go_term, (exp_direct, exp_all) in expected_results.items():
sys.stderr.write("Testing for term '{}' ...\n".format(go_term))
direct_anc, all_anc = mapslim(go_term, go_dag, goslim_dag)
if direct_anc != exp_direct or all_anc != exp_all:
tests_succeed = False
sys.stderr.write("failed.\n")
else:
sys.stderr.write("success!\n")
if tests_succeed:
print("All test passed successfully!")
sys.exit(0)
else:
sys.stderr.write("[ERROR] At least one test failed.\n")
sys.exit(1)
all_go_accs_in_a_protein = set()
all_goslim_anc_accs_in_a_protein = set()
all_goslim_covered_anc = set()
go_accs = set(interpro_go.loc[interpro_go['Protein Accession'] == protein]['GO Accession'])
for go_acc in go_accs:
if not pd.isnull(go_acc):
all_go_accs_in_a_protein |= set(go_acc.split('|'))
if len(all_go_accs_in_a_protein) > 0:
for go_term in all_go_accs_in_a_protein:
if go_term not in go:
continue
if USE_SLIM:
direct_anc, all_anc = mapslim(go_term, go, goslim)
all_goslim_anc_accs_in_a_protein |= all_anc
all_goslim_covered_anc |= (all_anc - direct_anc)
query_term = go.query_term(go_term)
output_table = output_table.append(pd.DataFrame({'Protein Accession': [protein], 'GO Category': [query_term.namespace], 'GO Accession': [go_term], 'GO Description': [query_term.name], 'GO Level':[query_term.level]}), ignore_index=True)
if USE_SLIM:
if ONLY_DIRECT:
all_goslim_direct_anc_accs_in_a_protein = all_goslim_anc_accs_in_a_protein - all_goslim_covered_anc
for goslim_term in all_goslim_direct_anc_accs_in_a_protein:
query_goslim_term = goslim.query_term(goslim_term)
output_slim_table = output_slim_table.append(pd.DataFrame({'Protein Accession': [protein], 'GO Category': [query_goslim_term.namespace], 'GOSlim Accession': [goslim_term], 'GOSlim Description': [query_goslim_term.name], 'GOSlim Level':[query_goslim_term.level]}), ignore_index=True)
else:
for goslim_term in all_goslim_anc_accs_in_a_protein:
query_goslim_term = goslim.query_term(goslim_term)
from goatools.mapslim import mapslim
from collections import Counter
go_dag = GODag('/home/gstupp/goatools/go.obo')
goslim_dag = GODag('/home/gstupp/goatools3/goslim_generic.obo')
goslim_meta = GODag('/home/gstupp/goatools3/goslim_metagenomics.obo')
ana_DTA = '/home/gstupp/01_2015_mass_spec/H1_11082014/1108_Gly1_2014_12_15_15_29205/dtaselect_results_sfp0.01_p2/DTASelect-filter.txt'
parser = blazmass_tools.dta_select_parser(ana_DTA, small=True)
ps = [get_domains(p) for p in parser]
set_go = set(chain(*[p['set_go'] for p in ps if p['set_go'] is not None]))
for p in ps:
if p['set_go']:
p['go_slim'] = set(
chain(*[
mapslim(go_term, go_dag, goslim_meta)[0]
for go_term in p['set_go'] if go_term in go_dag
]))
else:
p['go_slim'] = None
go_slim = Counter(chain(*[p['go_slim'] for p in ps if p['go_slim']]))
labels = {
go_term: go_dag.query_term(go_term).name
for go_term in go_slim.keys()
}
[labels[go] for (go, x) in go_slim.most_common(n=10)]
import plot_tools
cmap = plt.cm.jet
colors = cmap(np.linspace(0., 1., len(go_slim.keys())))