def get_geneid_to_localization(file_name, mapping_file, only_extracellular=False, filter_uncertain=False):
"""
Relevant extracellular term: Plasma membrane (GO:0005886)
Some other terms: Vesicles (GO:0043231);Nuclear bodies (GO:0016604);Nucleoplasm (GO:0005654);Cytosol (GO:0005829);Nucleoli (GO:0005730);Mitochondria (GO:0005739);Cell Junctions (GO:0030054);Endoplasmic reticulum (GO:0005783);Centrosome (GO:0005813);Nucleoli fibrillar center (GO:0001650);Golgi apparatus (GO:0005794);Nuclear speckles (GO:0016607)
Reliability types: Uncertain | Approved | Supported | Enhanced
"""
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(mapping_file)
geneid_to_localization = {}
ids_unmatched = set()
f = open(file_name)
f.readline()
for line in f:
words = line.strip("\n").split("\t")
# Evidence and reliability are for experimental characterization
gene, reliability, go_terms = words[1], words[2], words[-1]
if filter_uncertain and reliability == "Uncertain":
continue
if gene in name_to_geneid:
geneid = name_to_geneid[gene]
else:
#print "Unmatched id", gene
ids_unmatched.add(gene)
continue
for go in go_terms.split(";"):
idx = go.find("(")
go = go[idx+1:-1]
if only_extracellular and go != "GO:0005886":
continue
geneid_to_localization.setdefault(geneid, set()).add((go, reliability))
f.close()
print "Unmatched:", len(ids_unmatched) #, ", ".join(sorted(ids_unmatched))
return geneid_to_localization
def get_geneid_symbol_mapping(mapping_file):
"""
id_mapping_file = %(data_dir)s/ncbi/geneid_to_symbol.txt
"""
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(
mapping_file)
return geneid_to_name, name_to_geneid
def create_network_file(file_name, mapping_file, out_file, tissue=None):
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(
mapping_file)
parser = TsvReader.TsvReader(file_name, delim="\t")
fields_to_include = ["evidence type", "symbol1", "symbol2"]
if tissue is not None:
fields_to_include += [tissue]
column_to_index, id_to_values = parser.read(
fields_to_include=fields_to_include)
edges = set()
for evidence, values in id_to_values.iteritems():
if evidence == "pred":
continue
if tissue is not None:
for gene1, gene2, abundance in values:
#if "P" not in set(abundance):
if abundance == "-":
continue
if gene1 in name_to_geneid and gene2 in name_to_geneid:
geneid1 = name_to_geneid[gene1]
geneid2 = name_to_geneid[gene2]
edges.add((geneid1, geneid2))
else:
for gene1, gene2 in values:
if gene1 in name_to_geneid and gene2 in name_to_geneid:
geneid1 = name_to_geneid[gene1]
geneid2 = name_to_geneid[gene2]
edges.add((geneid1, geneid2))
f = open(out_file, 'w')
for u, v in edges:
f.write("%s 1 %s\n" % (u, v))
f.close()
return
def create_network_file(file_name, mapping_file, out_file, tissue=None):
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(mapping_file)
parser = TsvReader.TsvReader(file_name, delim="\t")
fields_to_include = ["evidence type", "symbol1", "symbol2"]
if tissue is not None:
fields_to_include += [tissue]
column_to_index, id_to_values = parser.read(fields_to_include=fields_to_include)
edges = set()
for evidence, values in id_to_values.iteritems():
if evidence == "pred":
continue
if tissue is not None:
for gene1, gene2, abundance in values:
#if "P" not in set(abundance):
if abundance == "-":
continue
if gene1 in name_to_geneid and gene2 in name_to_geneid:
geneid1 = name_to_geneid[gene1]
geneid2 = name_to_geneid[gene2]
edges.add((geneid1, geneid2))
else:
for gene1, gene2 in values:
if gene1 in name_to_geneid and gene2 in name_to_geneid:
geneid1 = name_to_geneid[gene1]
geneid2 = name_to_geneid[gene2]
edges.add((geneid1, geneid2))
f = open(out_file, 'w')
for u, v in edges:
f.write("%s 1 %s\n" % (u, v))
f.close()
return
def main():
base_dir = "/home/emre/arastirma/data/drug/sensitivity/gdsc/"
#file_name = base_dir + "gdsc_compounds_conc_w5.csv"
#compound_to_concentrations = get_compounds(file_name)
#print len(compound_to_concentrations), compound_to_concentrations["GNF-2"]
geneid_to_names, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(base_dir + "../../../proteome/ncbi/geneid_to_symbol.txt")
file_target = base_dir + "gdsc_en_output_w5.csv"
file_response = base_dir + "gdsc_manova_output_w5.csv"
get_gsdc_info(file_target, file_response, name_to_geneid)
return
def main():
mapping_file = "/home/emre/data/ncbi/geneid_to_symbol.txt"
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(mapping_file)
file_name = "/home/emre/data/tissue/hpa_subcellular_location.tsv"
geneid_to_localization = get_geneid_to_localization(file_name, mapping_file) # , only_extracellular=True
print geneid_to_localization["8648"] # NCOA1
print geneid_to_localization["207"] # AKT1
file_name_membrane = "/home/emre/data/tissue/hpa_predicted_membrane.tsv"
file_name_secreted = "/home/emre/data/tissue/hpa_predicted_secreted.tsv"
geneid_to_localization = get_geneid_to_predicted_localization(file_name_membrane, file_name_secreted, mapping_file)
print geneid_to_localization["5925"] # RB1
return
def get_geneid_to_predicted_localization(file_name_membrane, file_name_secreted, mapping_file):
"""
Relevant protein classes: Predicted secreted proteins | Predicted membrane proteins
Note that evidence and reliability are for experimental characterization
"""
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(mapping_file)
geneid_to_localization = {}
ids_unmatched = set()
f = open(file_name_membrane)
f.readline()
for line in f:
words = line.strip("\n").split("\t")
gene, protein_class, evidence, reliability = words[0], words[6], words[7], words[11]
#if reliability not in ("Approved", "Supported", "Enhanced"):
# continue
if gene in name_to_geneid:
geneid = name_to_geneid[gene]
else:
#print "Unmatched id", gene
ids_unmatched.add(gene)
continue
for localization in protein_class.split(", "):
localization = localization.replace("Predicted ", "").replace(" proteins", "").replace(" genes", "")
if localization == "membrane":
geneid_to_localization.setdefault(geneid, set()).add((localization, reliability))
f.close()
f = open(file_name_secreted)
f.readline()
for line in f:
words = line.strip("\n").split("\t")
# Evidence and reliability are for experimental characterization
gene, protein_class, evidence, reliability = words[0], words[6], words[7], words[11]
if gene in name_to_geneid:
geneid = name_to_geneid[gene]
else:
#print "Unmatched id", gene
ids_unmatched.add(gene)
continue
for localization in protein_class.split(", "):
localization = localization.replace("Predicted ", "").replace(" proteins", "").replace(" genes", "")
if localization == "secreted":
geneid_to_localization.setdefault(geneid, set()).add((localization, reliability))
f.close()
print "Unmatched:", len(ids_unmatched) #, ", ".join(sorted(ids_unmatched))
return geneid_to_localization
def get_geneid_symbol_mapping(mapping_file):
geneid_to_names, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(
mapping_file)
return geneid_to_names, name_to_geneid
def get_geneid_symbol_mapping(mapping_file):
"""
id_mapping_file = %(data_dir)s/ncbi/geneid_to_symbol.txt
"""
geneid_to_name, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(mapping_file)
return geneid_to_name, name_to_geneid
def get_geneid_symbol_mapping(mapping_file):
geneid_to_names, name_to_geneid = parse_ncbi.get_geneid_symbol_mapping(mapping_file)
return geneid_to_names, name_to_geneid
