def protein_item_statements():
"""
construct list of referenced statements to pass to PBB_Core Item engine
:return:
"""
uniprot_ref = wdo.reference_store(source='uniprot', identifier=uniprot)
WD_String_CLAIMS = {
'P637':
str(gene_record['refseq']['protein']), # set refseq protein id
'P352': uniprot # Set uniprot ID
}
WD_Item_CLAIMS = {
'P703': [spec_strain.iloc[0]['wd_qid']
], # get strain taxid qid from strain record
'P279': ['Q8054'], # subclass of protein
}
statements = []
#generate go term claims
for gt in gene_record['GOTERMS']:
goprop = go_props[gt[1]]
govalue = wdo.WDSparqlQueries(
prop='P686',
string=gt[0]).wd_prop2qid() # Get GeneOntology Item by GO ID
evprop = 'P459'
try:
evvalue = go_evidence_codes[gt[2]]
evstat = PBB_Core.WDItemID(value=evvalue,
prop_nr=evprop,
is_qualifier=True)
statements.append(
PBB_Core.WDItemID(value=govalue,
prop_nr=goprop,
references=[uniprot_ref],
qualifiers=[evstat]))
except Exception as e:
statements.append(
PBB_Core.WDItemID(value=govalue,
prop_nr=goprop,
references=[uniprot_ref]))
# generate list of pbb core value objects for all valid claims
for k, v in WD_Item_CLAIMS.items():
if v:
for i in v:
statements.append(
PBB_Core.WDItemID(value=i,
prop_nr=k,
references=[uniprot_ref]))
for k, v in WD_String_CLAIMS.items():
if v:
statements.append(
PBB_Core.WDString(value=v,
prop_nr=k,
references=[uniprot_ref]))
return statements
def gene_item_statements():
"""
construct list of referenced statements to past to PBB_Core Item engine
:return:
"""
# creates reference object for WD gene item claim
ncbi_gene_reference = wdo.reference_store(source='ncbi_gene', identifier=gene_record['_id'])
# claims for datatype string.
WD_String_CLAIMS = {'P351': str(gene_record['_id']),
'P2393': gene_record['locus_tag'],
'P644': str(int(gene_record['genomic_pos']['start'])),
'P645': str(int(gene_record['genomic_pos']['end'])),
}
# claims for datytpe item
WD_Item_CLAIMS = {'P703': spec_strain.iloc[0]['wd_qid'],
'P279': 'Q7187',
}
# convert integer representation of strand to corresponding WD item (Forward Strand/Reverse Strand)
if gene_record['genomic_pos']['strand'] == '1':
WD_Item_CLAIMS['P2548'] = 'Q22809680'
elif gene_record['genomic_pos']['strand'] == '-1':
WD_Item_CLAIMS['P2548'] = 'Q22809711'
statements = []
# process to pbb_Core data value object and append to statments for each valid item in each datatype dict
# WDItemID datatype
for k, v in WD_Item_CLAIMS.items():
statements.append(PBB_Core.WDItemID(value=v, prop_nr=k, references=[ncbi_gene_reference]))
# WDString datatype
for k, v in WD_String_CLAIMS.items():
statements.append(PBB_Core.WDString(value=v, prop_nr=k, references=[ncbi_gene_reference]))
return statements
def gene_item_statements():
"""
construct list of referenced statements to past to PBB_Core Item engine
:return:
"""
# creates reference object for WD gene item claim
ncbi_gene_reference = wdo.reference_store(
source='ncbi_gene', identifier=gene_record['_id'])
# claims for datatype string.
WD_String_CLAIMS = {
'P351': str(gene_record['_id']),
'P2393': gene_record['locus_tag'],
}
WD_Genome_Annotation_Claims = {
'P644': str(int(gene_record['genomic_pos']['start'])),
'P645': str(int(gene_record['genomic_pos']['end'])),
}
# claims for datytpe item
WD_Item_CLAIMS = {
'P703': spec_strain.iloc[0]['wd_qid'],
'P279': 'Q7187',
}
# convert integer representation of strand to corresponding WD item (Forward Strand/Reverse Strand)
if gene_record['genomic_pos']['strand'] == 1:
WD_Item_CLAIMS['P2548'] = 'Q22809680'
elif gene_record['genomic_pos']['strand'] == -1:
WD_Item_CLAIMS['P2548'] = 'Q22809711'
chromosome = gene_record['genomic_pos']['chr']
rs_chrom = PBB_Core.WDString(value=chromosome,
prop_nr='P2249',
is_qualifier=True)
statements = []
# process to pbb_Core data value object and append to statments for each valid item in each datatype dict
# WDItemID datatype
for k, v in WD_Item_CLAIMS.items():
statements.append(
PBB_Core.WDItemID(value=v,
prop_nr=k,
references=[ncbi_gene_reference]))
# WDString datatype
for k, v in WD_String_CLAIMS.items():
statements.append(
PBB_Core.WDString(value=v,
prop_nr=k,
references=[ncbi_gene_reference]))
for k, v in WD_Genome_Annotation_Claims.items():
statements.append(
PBB_Core.WDString(value=v,
prop_nr=k,
references=[ncbi_gene_reference],
qualifiers=[rs_chrom]))
return statements
def parse_go_terms(goclass):
"""
parses the tsv formated result from uniprot REST API
:param goclass:
:return:
"""
go_props = {'Function': 'P680', 'Component': 'P681', 'Process': 'P682'}
goqid = wdo.WDSparqlQueries(string=goclass[0],
prop='P686').wd_prop2qid()
return goqid, go_props[goclass[1]]
def protein_item_statements():
"""
construct list of referenced statements to past to PBB_Core Item engine
:return:
"""
uniprot_ref = wdo.reference_store(source='uniprot', identifier=uniprot)
WD_String_CLAIMS = {
'P637': str(gene_record['refseq']['protein']),
#'P2393': gene_record['locus_tag'],
'P352': uniprot
#'P591': str(gene_record['EC number'])
}
WD_Item_CLAIMS = {
'P703': [spec_strain.iloc[0]['wd_qid']],
'P279': ['Q8054'],
'P680': [], # molecular function
'P681': [], # cellular component
'P682': [] # biological process
}
for gt in gene_record['GOTERMS']:
gtids = parse_go_terms(gt)
WD_Item_CLAIMS[gtids[1]].append(gtids[0])
statements = []
# generate list of pbb core value objects for all valid claims
for k, v in WD_Item_CLAIMS.items():
if v:
for i in v:
statements.append(
PBB_Core.WDItemID(value=i,
prop_nr=k,
references=[uniprot_ref]))
for k, v in WD_String_CLAIMS.items():
if v:
statements.append(
PBB_Core.WDString(value=v,
prop_nr=k,
references=[uniprot_ref]))
return statements
def encodes(gene_record, login):
"""
identifies microbial gene and protein items and links them via encodes (P688) and encoded by (P702) functions
:param gene_record: gene record from MGI_UNIP_MERGER()
:return: links gene and protein wikidata items.
"""
uniprot = str(list(gene_record['uniprot'].values())[0])
start = time.time()
# find gene and protein qids
gene_qid = wdo.WDSparqlQueries(prop='P351',
string=gene_record['_id']).wd_prop2qid()
protein_qid = wdo.WDSparqlQueries(prop='P352',
string=uniprot).wd_prop2qid()
print(gene_qid, protein_qid)
# if a gene or protein item is not found skip this one
if gene_qid is not None and protein_qid is not None:
print('gene {} and protein {} found'.format(gene_qid, protein_qid))
# generate reference and claim values for each item
ncbi_gene_reference = wdo.reference_store(
source='ncbi_gene', identifier=gene_record['_id'])
gene_encodes = [
PBB_Core.WDItemID(value=protein_qid,
prop_nr='P688',
references=[ncbi_gene_reference])
]
protein_encoded_by = [
PBB_Core.WDItemID(value=gene_qid,
prop_nr='P702',
references=[ncbi_gene_reference])
]
# find and write items
success_count = 0
wd_encodes_item = PBB_Core.WDItemEngine(wd_item_id=gene_qid,
data=gene_encodes)
#pprint.pprint(wd_encodes_item.get_wd_json_representation())
try:
wd_encodes_item = PBB_Core.WDItemEngine(wd_item_id=gene_qid,
data=gene_encodes)
wd_encodes_item.write(login)
PBB_Core.WDItemEngine.log(
'INFO',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=gene_record['_id'],
exception_type='',
message='encodes claim written successfully',
wd_id=wd_encodes_item.wd_item_id,
duration=time.time() - start))
print('gene success')
success_count += 1
except Exception as e:
print(e)
PBB_Core.WDItemEngine.log(
'ERROR',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=gene_record['_id'],
exception_type=type(e),
message=e.__str__(),
wd_id='',
duration=time.time() - start))
try:
wd_encoded_by_item = PBB_Core.WDItemEngine(wd_item_id=protein_qid,
data=protein_encoded_by)
wd_encoded_by_item.write(login)
PBB_Core.WDItemEngine.log(
'INFO',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=uniprot,
exception_type='',
message='encoded by claim written successfully',
wd_id=wd_encoded_by_item.wd_item_id,
duration=time.time() - start))
print('protein success')
success_count += 1
except Exception as e:
print(e)
PBB_Core.WDItemEngine.log(
'ERROR',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=gene_record['_id'],
exception_type=type(e),
message=e.__str__(),
wd_id='',
duration=time.time() - start))
if success_count == 2:
return 'success'
end = time.time()
print('Time elapsed:', end - start)
def mol2qid(moldict):
mols = []
for key, value in moldict.items():
wd = wdo.WDSparqlQueries(prop='P683', string=value)
mols.append(wd.wd_prop2qid())
return [x for x in mols if x != 'None']
def encodes(gene_record, login):
"""
identifies microbial gene and protein items and links them via encodes (P688) and encoded by (P702) functions
:param gene_record: gene record from MGI_UNIP_MERGER()
:return: links gene and protein wikidata items.
"""
uniprot = str(list(gene_record['uniprot'].values())[0])
start = time.time()
# find gene and protein qids
gene_qid = wdo.WDSparqlQueries(prop='P351', string=gene_record['_id']).wd_prop2qid()
protein_qid = wdo.WDSparqlQueries(prop='P352', string=uniprot).wd_prop2qid()
# if a gene or protein item is not found skip this one
if gene_qid is not None and protein_qid is not None:
print('gene {} and protein {} found'.format(gene_qid, protein_qid))
# generate reference and claim values for each item
ncbi_gene_reference = wdo.reference_store(source='ncbi_gene', identifier=gene_record['_id'])
gene_encodes = [PBB_Core.WDItemID(value=protein_qid, prop_nr='P688', references=[ncbi_gene_reference])]
protein_encoded_by = [PBB_Core.WDItemID(value=gene_qid, prop_nr='P702', references=[ncbi_gene_reference])]
# find and write items
success_count = 0
try:
wd_encodes_item = PBB_Core.WDItemEngine(wd_item_id=gene_qid, data=gene_encodes)
wd_encodes_item.write(login)
PBB_Core.WDItemEngine.log('INFO', '{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'.format(
main_data_id=gene_record['_id'],
exception_type='',
message='encodes claim written successfully',
wd_id=wd_encodes_item.wd_item_id,
duration=time.time() - start
)
)
print('gene success')
success_count += 1
except Exception as e:
print(e)
PBB_Core.WDItemEngine.log('ERROR', '{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'.format(
main_data_id=gene_record['_id'],
exception_type=type(e),
message=e.__str__(),
wd_id='',
duration=time.time() - start
))
try:
wd_encoded_by_item = PBB_Core.WDItemEngine(wd_item_id=protein_qid, data=protein_encoded_by)
wd_encoded_by_item.write(login)
PBB_Core.WDItemEngine.log('INFO', '{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'.format(
main_data_id=uniprot,
exception_type='',
message='encoded by claim written successfully',
wd_id=wd_encoded_by_item.wd_item_id,
duration=time.time() - start
)
)
print('protein success')
success_count += 1
except Exception as e:
print(e)
PBB_Core.WDItemEngine.log('ERROR', '{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'.format(
main_data_id=gene_record['_id'],
exception_type=type(e),
message=e.__str__(),
wd_id='',
duration=time.time() - start
))
if success_count == 2:
return 'success'
end = time.time()
print('Time elapsed:', end - start)
def sparql_qid(taxid):
qidobj = wdo.WDSparqlQueries(string=taxid, prop='P685')
return qidobj.wd_prop2qid()