def load_syngo_genes():
syngo = Ontology.from_table('../prev_databases/SynGO_BP.txt')
syngo_bp_genes = syngo.genes
syngo = Ontology.from_table('../prev_databases/SynGO_CC.txt')
syngo_cc_genes = syngo.genes
syngo_genes = list(set(syngo_bp_genes + syngo_cc_genes))
return syngo_genes
def load_syngo_genes():
syngo=Ontology.from_table('/Users/karenmei/Documents/Synapse_Ontology/NetworkClass/Metrics/SynGO_BP.txt')
syngo_bp_genes=syngo.genes
syngo=Ontology.from_table('/Users/karenmei/Documents/Synapse_Ontology/NetworkClass/Metrics/SynGO_CC.txt')
syngo_cc_genes=syngo.genes
syngo_genes=list(set(syngo_bp_genes+syngo_cc_genes))
training=load_training_genes()
syngo_genes=list(set(syngo_genes)-set(training))
return syngo_genes
def compare(ont):
#Generate custom ontology 1 from synapse branch from human GO
#ontology_file=Generate_Ontology_File('GO:0045202', 'ont1.txt')
ont1=Ontology.from_table('../../analyses/Histogram_ont/synapse.txt')
#print ('num synapse ontology terms', len(ont1.terms))
ont2=Ontology.from_table(fn)
#print ('num custom ontology terms', len(ont2.terms))
return(Num_Enriched_Modules(ont1, ont2))
def compare_to_go(ont_fn):
#Generate custom ontology 1 from synapse branch from human GO
#ontology_file=Generate_Ontology_File('GO:0045202', 'ont1.txt')
synapse_ont = '/home/joreyna/projects/BNFO286/synapse_ontology/analyses/Histogram_ont/synapse.txt'
ont1 = Ontology.from_table(synapse_ont)
#print ('num synapse ontology terms', len(ont1.terms))
ont2 = Ontology.from_table(ont_fn)
#print ('num custom ontology terms', len(ont2.terms))
return (Num_Enriched_Modules(ont1, ont2))
def create_ddot_object(self, filename="ddot_output.txt", **kwargs):
from ddot import Ontology
self.ontology = Ontology.from_table(filename,
clixo_format=True,
**kwargs)
return self
def load_pheno_hpo():
hpo = Ontology.from_table(
'/Users/karenmei/Documents/Synapse_Ontology/HPO/making_HPO/HPO_parent_child.txt'
)
#print (hpo)
hpo = hpo.propagate(direction='forward', gene_term=True, term_term=False)
hpo = hpo.focus(branches=['Phenotypic abnormality'])
return hpo
def run_ddot(theargs):
try:
(e_code, c_out, c_err) = run_clixo(theargs.clixopath, theargs.input,
theargs.alpha, theargs.beta)
df = pd.read_csv(io.StringIO(c_out.decode('utf-8')),
sep='\t',
engine='python',
header=None,
comment='#')
if theargs.output is not None:
try:
with open(theargs.output, 'wb') as f:
f.write(c_out)
outputdir = os.path.dirname(theargs.output)
statres = os.stat(outputdir)
os.chown(theargs.output, statres[stat.ST_UID],
statres[stat.ST_GID])
except Exception as ex:
sys.stderr.write('Caught exception trying to write file or'
'change permission: ' + str(ex))
ont1 = Ontology.from_table(df, clixo_format=True, parent=0, child=1)
if theargs.ndexserver.startswith('http://'):
server = theargs.ndexserver
else:
server = 'http://' + theargs.ndexserver
idf = pd.read_csv(theargs.input,
sep='\t',
engine='python',
header=None,
comment='#')
idf.rename(columns={
0: 'Gene1',
1: 'Gene2',
2: 'has_edge'
},
inplace=True)
ont_url, G = ont1.to_ndex(name=theargs.ndexname,
network=idf,
main_feature='has_edge',
ndex_server=server,
ndex_pass=theargs.ndexuser,
ndex_user=theargs.ndexpass,
layout=theargs.ndexlayout,
visibility=theargs.ndexvisibility)
return 'RESULT:' + ont_url.strip().replace('/v2/network/',
'/#/network/') + '\n'
except OverflowError as ofe:
logger.exception('Error running clixo')
return 'ERROR:' + str(ofe) + '\n'
except Exception as e:
logger.exception('Some other error')
return 'ERROR:' + str(e) + '\n'
return {'error': 'unknown error'}
def load_syngo_presynapse():
syngo = Ontology.from_table(
'/Users/karenmei/Documents/Synapse_Ontology/NetworkClass/Metrics/SynGO_CC.txt'
)
syngo = syngo.propagate(direction='forward',
gene_term=True,
term_term=False)
pre = syngo.focus(branches=['presynapse'])
pre_genes = pre.genes
return pre_genes
def upload_file():
#=================
# default values
#=================
alpha = 0.05
beta = 0.5
try:
data = request.files.get('file')
except Exception as e:
raise HTTPError(500, e)
if data and data.file:
if (request.query.alpha):
alpha = request.query.alpha
if (request.query.beta):
beta = request.query.beta
with tempfile.NamedTemporaryFile('w', delete=False) as f:
f.write(data.file.read())
f_name = f.name
f.close()
try:
clixo_file = generate_clixo_file(f_name, alpha, beta)
with open(clixo_file, 'r') as f_saved:
df = pd.read_csv(f_saved, sep='\t', engine='python', header=None, comment='#')
print(df.columns)
ont1 = Ontology.from_table(df, clixo_format=True, parent=0, child=1)
ont_url, G = ont1.to_ndex(name='MODY',
ndex_server='http://test.ndexbio.org',
ndex_pass='******',
ndex_user='******',
layout='bubble-collect',
visibility='PUBLIC')
if ont_url is not None and len(ont_url) > 0 and 'http' in ont_url:
uuid = ont_url.split('/')[-1]
return 'File has been processed. UUID:: %s \n' % uuid
else:
return 'File has been processed. UUID: %s \n' % ont_url
print('File has been processed: %s' % ont_url)
except OverflowError as ofe:
print('Error with running clixo')
else:
raise HTTPError(422, '**** FILE IS MISSING ****')
return "Unable to complete process. See stack message above."
def analyze_gene_enrichment(ont_fn, disease_gene_fn):
## We read in our ontology
#Generate custom ontology from Chromatin branch from human GO
#ontology_file=Generate_Ontology_File('GO:0000785')
ont = Ontology.from_table(ont_fn)
translated = Find_GO_Focus_GeneDict(ont)
#Test genes: autism
text_file = open(disease_gene_fn, "r")
test_gene_list = text_file.read().splitlines()
text_file.close()
#print("Number of autism genes in our ontology:" , len(set(ont.genes).intersection(set(test_gene_list))))
num_ont_disease_genes = len(
set(ont.genes).intersection(set(test_gene_list)))
#Find number of test genes in enriched modules:
num_enriched_disease_genes = Find_num_genes_in_enriched(
ont, translated, test_gene_list)
return (num_ont_disease_genes, num_enriched_disease_genes)
def metric_1(ont_file, test_gene_list):
ont1 = Ontology.from_table(ont_file)
ont1_genes = ont1.genes
test_recovery = jaccard(ont1_genes, test_gene_list)
print('recovery of test genes:', test_recovery)
return test_recovery
def load_pheno_hpo():
hpo=Ontology.from_table('../other_resources/HPO_parent_child.txt')
#print (hpo)
hpo=hpo.propagate(direction='forward', gene_term=True, term_term=False)
hpo=hpo.focus(branches=['Phenotypic abnormality'])
return hpo
true_terms=Find_Enrichment(ont, chr_ont, test_gene_list)
genes_in_true_terms=[]
for item in true_terms:
term=item[0]
genes=chr_ont[term]
genes_in_true_terms.append(genes)
print ('genes in true terms:', genes_in_true_terms)
overlap=list(set(genes_in_true_terms[0])&set(test_gene_list))
overlap_num=len(overlap)
print (overlap_num)
return overlap_num
#Test genes: a list of some genes involved in transcriptionally active chromatin
test_gene_list=['AFF4', 'PSIP1', 'H2AFB1', 'H2AFB2', 'H2AFB3', 'TTC37', 'WDR61', 'KMT2E', 'BCAS3', 'HIST1H1C', 'ZC3H8', 'CTR9', 'PADI2', 'PAF1', 'SUPT6H', 'EXOSC4', 'ELL', 'PCID2', 'EXOSC5', 'PELP1', 'ESR1', 'EXOSC10', 'EXOSC3', 'ICE1', 'ICE2']
test_gene_list = 'AAK1 ABCC8 ABHD17A ABHD17B ABHD17C ABHD6 ABI1 ABI2 ABL1'.split()
#Generate custom ontology from Chromatin branch from human GO
#ontology_file=Generate_Ontology_File('GO:0000785')
ont=Ontology.from_table('./ont2.txt')
#Make dictionary of terms and genes within custom ontology
chr_ont=Find_GO_Focus_GeneDict(ont)
#Find number of test genes in enriched modules:
Find_num_genes_in_enriched(ont, chr_ont, test_gene_list)
if len(test_gene_list) != 0:
enriched_terms = Find_Enrichment(ont1, chr_ont, test_gene_list)
enriched_modules.append(enriched_terms)
else:
continue
enriched_modules = [x for x in enriched_modules if x != []]
enriched_modules = [
item for sublist in enriched_modules for item in sublist
]
enriched_term_names = []
for item in enriched_modules:
enriched_term_name = item[0]
enriched_term_names.append(enriched_term_name)
unique_enriched_term_names = set(enriched_term_names)
#print (unique_enriched_term_names)
print('Num of Enriched Modules', len(unique_enriched_term_names))
return
#Generate custom ontology 1 from Chromatin branch from human GO
ontology_file = Generate_Ontology_File('GO:0000785', 'ont1.txt')
ont1 = Ontology.from_table('ont1.txt')
#print (ont1)
#Generate custom ontology 2 from Chromosome branch from human GO
ont2_file = Generate_Ontology_File('GO:0005694', 'ont2.txt')
ont2 = Ontology.from_table('ont2.txt')
#print (ont2)
Num_Enriched_Modules(ont1, ont2)
# Download gene-term annotations for human
r = requests.get('http://geneontology.org/gene-associations/goa_human.gaf.gz')
with open(path + 'goa_human.gaf.gz', 'wb') as f:
f.write(r.content)
hierarchy = pd.read_table('go.tab',
sep='\t',
header=None,
names=['Parent', 'Child', 'Relation', 'Namespace'])
with gzip.open(path + 'goa_human.gaf.gz', 'rb') as f:
mapping = ddot.parse_gaf(f)
ontGO = Ontology.from_table(table=hierarchy,
parent='Parent',
child='Child',
mapping=mapping,
mapping_child='DB Object ID',
mapping_parent='GO ID',
add_root_name='GO:00SUPER',
ignore_orphan_terms=True)
ontGO.clear_node_attr()
ontGO.clear_edge_attr()
go_descriptions = pd.read_table('goID_2_name.tab',
header=None,
names=['Term', 'Term_Description'],
index_col=0)
ontGO.update_node_attr(go_descriptions)
ontGO = ontGO.collapse_ontology(method='mhkramer')
if 'GO:00SUPER' not in ontGO.terms: ontGO.add_root('GO:00SUPER', inplace=True)
genes_in_true_terms = []
for item in true_terms:
term = item[0]
genes = chr_ont[term]
genes_in_true_terms.append(genes)
print('genes in true terms:', genes_in_true_terms)
overlap = list(set(genes_in_true_terms[0]) & set(test_gene_list))
overlap_num = len(overlap)
print(overlap_num)
return overlap_num
#Test genes: a list of some genes involved in transcriptionally active chromatin
test_gene_list = [
'AFF4', 'PSIP1', 'H2AFB1', 'H2AFB2', 'H2AFB3', 'TTC37', 'WDR61', 'KMT2E',
'BCAS3', 'HIST1H1C', 'ZC3H8', 'CTR9', 'PADI2', 'PAF1', 'SUPT6H', 'EXOSC4',
'ELL', 'PCID2', 'EXOSC5', 'PELP1', 'ESR1', 'EXOSC10', 'EXOSC3', 'ICE1',
'ICE2'
]
#Generate custom ontology from Chromatin branch from human GO
ontology_file = Generate_Ontology_File('GO:0000785')
ont = Ontology.from_table('custom_ontology.txt')
#Make dictionary of terms and genes within custom ontology
chr_ont = Find_GO_Focus_GeneDict(ont)
#Find number of test genes in enriched modules:
Find_num_genes_in_enriched(ont, chr_ont, test_gene_list)
uniq_parents = sorted(clixo_ont['Parent'].unique())
for parent in uniq_parents:
fw.write('{}\t{}\n'.format(parent, parent))
#
# for line in f:
# if line.startswith('#'):
# continue
#
# header = line.split()[0:3]
# parents = set()
# for line in f:
# line = line.strip().split('\t')
# parents.add(line[0])
#
# for parent in sorted(parents):
# fw.write('{}\t{}'.format(parent, parent))
with open(genes_fn) as f:
test_gene_list = [x.strip() for x in f.readlines()]
ont = Ontology.from_table(mod_ont_fn)
# Make dictionary of terms and genes within custom ontology
chr_ont = Find_GO_Focus_GeneDict(ont)
#print(chr_ont)
# Find number of test genes in enriched modules:
Find_num_genes_in_enriched(ont, chr_ont, test_gene_list)