def returnDirectories(sub_dir):
dir_list = unique.returnDirectories(sub_dir)
return dir_list
def exportTables(metabolite_list):
infer_enzyme_to_metabolite_pathway_data = 'no'
current_species_dirs = unique.returnDirectories('/Databases')
### Save results to all species directories
for species_code in current_species_dirs:
print 'Exporting metabolite data for:', species_code
gene_dir = 'Databases/' + species_code + '/gene/HMDB.txt'
gene_data = export.ExportFile(gene_dir)
hmdb_cas_dir = 'Databases/' + species_code + '/uid-gene/HMDB-CAS.txt'
hmdb_cas_data = export.ExportFile(hmdb_cas_dir)
hmdb_chebi_dir = 'Databases/' + species_code + '/uid-gene/HMDB-ChEBI.txt'
hmdb_chebi_data = export.ExportFile(hmdb_chebi_dir)
hmdb_pubchem_dir = 'Databases/' + species_code + '/uid-gene/HMDB-PubChem.txt'
hmdb_pubchem_data = export.ExportFile(hmdb_pubchem_dir)
hmdb_keggcomp_dir = 'Databases/' + species_code + '/uid-gene/HMDB-KeggCompound.txt'
hmdb_keggcomp_data = export.ExportFile(hmdb_keggcomp_dir)
hmdb_mapp_dir = 'Databases/' + species_code + '/gene-mapp/HMDB-MAPP.txt'
hmdb_mapp_data = export.ExportFile(hmdb_mapp_dir)
cas_denom_dir = 'Databases/' + species_code + '/gene-mapp/denominator/CAS.txt'
cas_denom_data = export.ExportFile(cas_denom_dir)
hmdb_go_dir = 'Databases/' + species_code + '/gene-go/HMDB-GeneOntology.txt'
if infer_enzyme_to_metabolite_pathway_data == 'yes':
hmdb_go_data = export.ExportFile(hmdb_go_dir)
headers = [
'hmdb_id', 'name', 'description', 'secondary_id', 'iupac',
'cas_number', 'chebi_id', 'pubchem_compound_id', 'Pathways',
'ProteinNames'
]
headers = string.join(headers, '\t') + '\n'
gene_data.write(headers)
### Attempt to add GO and pathway data from database based on associated protein IDs (simple translation from human)
mod = 'Ensembl'
try:
gene_annotations = gene_associations.importGeneData(
species_code, mod)
except Exception:
mod = 'EntrezGene'
try:
gene_annotations = gene_associations.importGeneData(
species_code, mod)
except Exception:
gene_annotations = {}
symbol_associations = {}
for geneid in gene_annotations:
symbol_associations[
gene_annotations[geneid].SymbolLower()] = geneid
gotype = 'null'
try:
gene_to_go = gene_associations.importGeneGOData(
species_code, mod, gotype)
except Exception:
gene_to_go = {}
try:
gene_to_mapp = gene_associations.importGeneMAPPData(
species_code, mod)
except Exception:
gene_to_mapp = {}
for ed in metabolite_list:
values = [
ed.HMDB(),
ed.Name(),
ed.Description(),
ed.SecondaryIDs(),
ed.IUPAC(),
ed.CAS(),
ed.CheBI(),
ed.PubChem(),
ed.PathwaysStr(),
ed.ProteinNamesStr()
]
values = string.join(values, '\t') + '\n'
gene_data.write(values)
if len(ed.Pathways()) > 1:
for pathway in ed.Pathways():
values = [ed.HMDB(), 'Ch', pathway]
values = string.join(values, '\t') + '\n'
hmdb_mapp_data.write(values)
if len(ed.CAS()) > 0:
values = [ed.HMDB(), ed.CAS()]
values = string.join(values, '\t') + '\n'
hmdb_cas_data.write(values)
values = [ed.CAS(), 'Ca']
values = string.join(values, '\t') + '\n'
cas_denom_data.write(values)
if len(ed.CheBI()) > 0:
values = [ed.HMDB(), ed.CheBI()]
values = string.join(values, '\t') + '\n'
hmdb_chebi_data.write(values)
if len(ed.PubChem()) > 0:
values = [ed.HMDB(), ed.PubChem()]
values = string.join(values, '\t') + '\n'
hmdb_pubchem_data.write(values)
if len(ed.KEGGCompoundID()) > 0:
values = [ed.HMDB(), ed.KEGGCompoundID()]
values = string.join(values, '\t') + '\n'
hmdb_keggcomp_data.write(values)
temp_go = {}
temp_mapp = {}
if infer_enzyme_to_metabolite_pathway_data == 'yes':
### If associated enzyme annotated, use the gene symbol to find GO terms associated with the gene symbol for the metabolite
### Not sure if this is a bad idea or not
for protein_name in ed.ProteinNames():
protein_name = string.lower(protein_name)
if protein_name in symbol_associations:
geneid = symbol_associations[protein_name]
if geneid in gene_to_go:
for goid in gene_to_go[geneid]:
temp_go[goid] = []
if geneid in gene_to_mapp:
for mapp in gene_to_mapp[geneid]:
temp_mapp[mapp] = []
for goid in temp_go:
values = [ed.HMDB(), 'GO:' + goid]
values = string.join(values, '\t') + '\n'
hmdb_go_data.write(values)
for mapp in temp_mapp:
values = [ed.HMDB(), 'Ch', mapp]
values = string.join(values, '\t') + '\n'
hmdb_mapp_data.write(values)
gene_data.close()
hmdb_mapp_data.close()
hmdb_cas_data.close()
hmdb_chebi_data.close()
hmdb_pubchem_data.close()
if infer_enzyme_to_metabolite_pathway_data == 'yes':
hmdb_go_data.close()
print 'File:', gene_dir, 'exported.'
def returnDirectories(sub_dir):
dir_list = unique.returnDirectories(sub_dir)
return dir_list