def create_reference(self):
"""
Create wikidata references for interpro
Items:
Q3047275: InterPro
Properties:
stated in (P248)
imported from (P143)
software version (P348)
publication date (P577)
"""
# This same reference will be used for everything. Except for a ref to the interpro item itself
ref_stated_in = PBB_Core.WDItemID("Q3047275",
'P248',
is_reference=True)
ref_imported = PBB_Core.WDItemID("Q3047275", 'P143', is_reference=True)
ref_version = PBB_Core.WDString(self.version,
'P348',
is_reference=True)
ref_date = PBB_Core.WDTime(self.date.strftime("+%Y-%m-%dT00:00:00Z"),
'P577',
is_reference=True)
ref_ipr = PBB_Core.WDString(self.id, "P2926", is_reference=True)
self.reference = [
ref_stated_in, ref_imported, ref_version, ref_date, ref_ipr
]
for ref in self.reference:
ref.overwrite_references = True
def make_ref_source(source_doc, id_prop, identifier, login=None):
"""
Reference is made up of:
stated_in: if the source has a release #:
release edition
else, stated in the source
link to id: link to identifier in source
retrieved: only if source has no release #
login: must be passed if you want to be able to create new release items
:param source_doc:
:param id_prop:
:param identifier:
:return:
"""
# source_doc = {'_id': 'uniprot', 'timestamp': '20161006'}
# source_doc = {'_id': 'ensembl', 'release': 86, 'timestamp': '20161005'}
source = source_doc['_id']
if source not in source_items:
raise ValueError(
"Unknown source for reference creation: {}".format(source))
if id_prop not in prop_ids:
raise ValueError(
"Unknown id_prop for reference creation: {}".format(id_prop))
link_to_id = PBB_Core.WDString(value=str(identifier),
prop_nr=prop_ids[id_prop],
is_reference=True)
if "release" in source_doc:
source_doc['release'] = str(source_doc['release'])
title = "{} Release {}".format(source_doc['_id'],
source_doc['release'])
description = "Release {} of {}".format(source_doc['release'],
source_doc['_id'])
edition_of_wdid = source_items[source_doc['_id']]
release = PBB_Helpers.Release(
title,
description,
source_doc['release'],
edition_of_wdid=edition_of_wdid).get_or_create(login)
stated_in = PBB_Core.WDItemID(value=release,
prop_nr='P248',
is_reference=True)
reference = [stated_in, link_to_id]
else:
date_string = source_doc['timestamp']
retrieved = datetime.strptime(date_string, "%Y%m%d")
stated_in = PBB_Core.WDItemID(value=source_items[source],
prop_nr='P248',
is_reference=True)
retrieved = PBB_Core.WDTime(retrieved.strftime('+%Y-%m-%dT00:00:00Z'),
prop_nr='P813',
is_reference=True)
reference = [stated_in, retrieved, link_to_id]
return reference
def create_relationships(self, ipr_wd):
# ipr_wd is a dict ipr ID to wikidata ID mapping
statements = [
PBB_Core.WDExternalID(value=self.id,
prop_nr=INTERPRO,
references=[self.reference])
]
if self.parent:
statements.append(
PBB_Core.WDItemID(value=ipr_wd[self.parent],
prop_nr='P279',
references=[self.reference])) # subclass of
if self.contains:
for c in self.contains:
statements.append(
PBB_Core.WDItemID(value=ipr_wd[c],
prop_nr='P527',
references=[self.reference])) # has part
if self.found_in:
for f in self.found_in:
statements.append(
PBB_Core.WDItemID(value=ipr_wd[f],
prop_nr='P361',
references=[self.reference])) # part of
if len(statements) == 1:
return
# write data
item = PBB_Core.WDItemEngine(item_name=self.name,
domain='interpro',
data=statements,
server=SERVER,
append_value=["P279", "P527", "P361"])
try:
item.write(self.login)
except WDApiError as e:
print(e)
PBB_Core.WDItemEngine.log(
'ERROR',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=self.id,
exception_type=type(e),
message=e.__str__(),
wd_id=self.wd_item_id,
duration=datetime.now()))
return
PBB_Core.WDItemEngine.log(
'INFO',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=self.id,
exception_type='',
message='created interpro relationships: {}'.format([
(x.prop_nr, x.value) for x in statements
]),
wd_id=item.wd_item_id,
duration=datetime.now()))
def create_relationships(self, login):
try:
# endpoint may not get updated in time?
self.do_wdid_lookup()
except KeyError as e:
PBB_Core.WDItemEngine.log(
"ERROR", format_msg(self.id, INTERPRO, None, str(e), type(e)))
return
statements = [
PBB_Core.WDExternalID(value=self.id,
prop_nr=INTERPRO,
references=[self.reference])
]
if self.parent:
# subclass of
statements.append(
PBB_Core.WDItemID(value=self.parent_wdid,
prop_nr='P279',
references=[self.reference]))
if self.contains:
for c in self.contains_wdid:
statements.append(
PBB_Core.WDItemID(value=c,
prop_nr='P527',
references=[self.reference])) # has part
if self.found_in:
for f in self.found_in_wdid:
statements.append(
PBB_Core.WDItemID(value=f,
prop_nr='P361',
references=[self.reference])) # part of
if len(statements) == 1:
return
wd_item = PBB_Core.WDItemEngine(
wd_item_id=self.wdid,
domain='interpro',
data=statements,
append_value=['P279', 'P527', 'P361'],
fast_run=True,
fast_run_base_filter=IPRTerm.fast_run_base_filter)
PBB_Helpers.try_write(
wd_item,
self.id,
INTERPRO,
login,
edit_summary="create/update subclass/has part/part of")
def gene_encodes_statement(gene_qid, protein_qid, id_prop, external_id, source,
login):
"""
:param gene_qid:
:param protein_qid:
:param id_prop:
:param external_id:
:param source:
:param login:
:return:
"""
ensembl_protein_reference = make_ref_source(source, id_prop, external_id)
# gene
gene_encodes = PBB_Core.WDItemID(value=protein_qid,
prop_nr='P688',
references=[ensembl_protein_reference])
wd_item_protein = PBB_Core.WDItemEngine(wd_item_id=gene_qid,
domain='genes',
data=[gene_encodes],
fast_run=True,
fast_run_base_filter={
'P351':
'',
'P703':
strain_info['organism_wdid']
})
if wd_item_protein.create_new_item:
raise ValueError("nooo!!")
try_write(wd_item_protein, external_id, id_prop, login)
def create_item(self, login):
statements = [
PBB_Core.WDExternalID(value=self.id,
prop_nr=INTERPRO,
references=[self.reference]),
PBB_Core.WDItemID(value=self.type_wdid,
prop_nr="P279",
references=[self.reference])
]
wd_item = PBB_Core.WDItemEngine(
item_name=self.name,
domain='interpro',
data=statements,
append_value=["P279"],
fast_run=True,
fast_run_base_filter=IPRTerm.fast_run_base_filter)
wd_item.set_label(self.name, lang='en')
for lang, description in self.lang_descr.items():
wd_item.set_description(description, lang=lang)
wd_item.set_aliases([self.short_name, self.id])
PBB_Helpers.try_write(wd_item, self.id, INTERPRO, login)
return wd_item
def create_reference(self):
ref_stated_in = PBB_Core.WDItemID(self.MONDO_WDID, 'P248', is_reference=True)
ref_retrieved = PBB_Core.WDTime(self.retrieved.strftime('+%Y-%m-%dT00:00:00Z'), 'P813',
is_reference=True) # interpro ID
#ref_archive_url = PBB_Core.WDUrl(self.ref_url, 'P1065', is_reference=True)
#reference = [ref_stated_in, ref_retrieved, ref_archive_url]
reference = [ref_stated_in, ref_retrieved]
self.reference = reference
def create_uniprot_relationships(login, release_wdid, collection, taxon=None):
# only do uniprot proteins that are already in wikidata
if taxon:
uniprot2wd = PBB_Helpers.id_mapper(UNIPROT, (("P703", taxon),))
fast_run_base_filter = {UNIPROT: "", "P703": taxon}
else:
uniprot2wd = PBB_Helpers.id_mapper(UNIPROT)
fast_run_base_filter = {UNIPROT: ""}
cursor = collection.find({'_id': {'$in': list(uniprot2wd.keys())}}, no_cursor_timeout=True)
for doc in tqdm(cursor, total=cursor.count()):
uniprot_id = doc['_id']
statements = []
# uniprot ID. needed for PBB_core to find uniprot item
# statements.append(PBB_Core.WDExternalID(value=uniprot_id, prop_nr=UNIPROT))
## References
# stated in Interpro version XX.X
ref_stated_in = PBB_Core.WDItemID(release_wdid, 'P248', is_reference=True)
ref_ipr = PBB_Core.WDString("http://www.ebi.ac.uk/interpro/protein/{}".format(uniprot_id), "P854",
is_reference=True)
reference = [ref_stated_in, ref_ipr]
if doc['subclass']:
for f in doc['subclass']:
statements.append(PBB_Core.WDItemID(value=IPRTerm.ipr2wd[f], prop_nr='P279', references=[reference]))
if doc['has_part']:
for hp in doc['has_part']:
statements.append(PBB_Core.WDItemID(value=IPRTerm.ipr2wd[hp], prop_nr='P527', references=[reference]))
if uniprot_id not in uniprot2wd:
print("wdid_not_found " + uniprot_id + " " + uniprot2wd[uniprot_id])
PBB_Core.WDItemEngine.log("ERROR", PBB_Helpers.format_msg(uniprot_id, UNIPROT, None, "wdid_not_found"))
wd_item = PBB_Core.WDItemEngine(wd_item_id=uniprot2wd[uniprot_id], domain="proteins", data=statements,
fast_run=True, fast_run_base_filter=fast_run_base_filter,
append_value=["P279", "P527", "P361"])
if wd_item.create_new_item:
raise ValueError("something bad happened")
PBB_Helpers.try_write(wd_item, uniprot_id, INTERPRO, login, edit_summary="add/update family and/or domains")
cursor.close()
def create_reference(self):
""" Create wikidata references for interpro
This same reference will be used for everything. Except for a ref to the interpro item itself
"""
# stated in Interpro version XX.X
ref_stated_in = PBB_Core.WDItemID(self.release_wdid,
'P248',
is_reference=True)
ref_ipr = PBB_Core.WDString(self.id, INTERPRO,
is_reference=True) # interpro ID
self.reference = [ref_stated_in, ref_ipr]
def make_ref(retrieved, genome_id):
refs = [
PBB_Core.WDItemID(value='Q20641742', prop_nr='P248',
is_reference=True), # stated in ncbi gene
PBB_Core.WDString(value=genome_id, prop_nr='P2249',
is_reference=True), # Link to Refseq Genome ID
PBB_Core.WDTime(retrieved.strftime('+%Y-%m-%dT00:00:00Z'),
prop_nr='P813',
is_reference=True)
]
return refs
def make_reference(source, id_prop, identifier, retrieved):
reference = [
PBB_Core.WDItemID(value=source_items[source],
prop_nr='P248',
is_reference=True), # stated in
PBB_Core.WDString(value=str(identifier),
prop_nr=prop_ids[id_prop],
is_reference=True), # Link to ID
PBB_Core.WDTime(retrieved.strftime('+%Y-%m-%dT00:00:00Z'),
prop_nr='P813',
is_reference=True)
]
return reference
def create_item(self):
statements = [
PBB_Core.WDExternalID(value=self.id,
prop_nr=INTERPRO,
references=[self.reference]),
PBB_Core.WDItemID(value=IPRItem.type2subclass[self.type],
prop_nr="P279",
references=[self.reference])
]
item = PBB_Core.WDItemEngine(item_name=self.name,
domain='interpro',
data=statements,
server=SERVER)
item.set_label(self.name)
for lang, description in self.description.items():
item.set_description(description, lang=lang)
item.set_aliases([self.short_name, self.id])
try:
item.write(login=self.login)
except WDApiError as e:
print(e)
PBB_Core.WDItemEngine.log(
'ERROR',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=self.id,
exception_type=type(e),
message=e.__str__(),
wd_id=self.wd_item_id,
duration=datetime.now()))
return
self.wd_item_id = item.wd_item_id
PBB_Core.WDItemEngine.log(
'INFO',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=self.id,
exception_type='',
message='created/updated interpro item',
wd_id=item.wd_item_id,
duration=datetime.now()))
def protein_item(record, strain_info, gene_qid, go_wdid_mapping, login,
add_pubmed):
"""
generate pbb_core item object
"""
item_name = '{} {}'.format(record['name']['@value'],
record['ensembl']['@value']['protein'])
item_description = '{} protein found in {}'.format(
strain_info['organism_type'], strain_info['organism_name'])
s = []
############
# external IDs
############
# will be used for reference statements
external_ids = {
'entrez_gene': str(record['entrezgene']['@value']),
'ensembl_protein': record['ensembl']['@value']['protein'],
'ensembl_gene': record['ensembl']['@value']['gene'],
'refseq_protein': record['refseq']['@value']['protein'],
'uniprot': record['uniprot']['@value']['Swiss-Prot']
}
# ensembl protein id
ensembl_ref = make_ref_source(record['ensembl']['@source'],
'ensembl_protein',
external_ids['ensembl_protein'])
s.append(
PBB_Core.WDString(external_ids['ensembl_protein'],
'P705',
references=[ensembl_ref]))
# refseq protein id
refseq_ref = make_ref_source(record['refseq']['@source'], 'refseq_protein',
external_ids['refseq_protein'])
s.append(
PBB_Core.WDString(external_ids['refseq_protein'],
'P637',
references=[refseq_ref]))
# uniprot id
uniprot_ref = make_ref_source(record['uniprot']['@source'], 'uniprot',
external_ids['uniprot'])
s.append(
PBB_Core.WDString(external_ids['uniprot'],
'P352',
references=[uniprot_ref]))
############
# GO terms
# TODO: https://www.wikidata.org/wiki/Q3460832
############
preprocess_go(record)
print(record)
go_source = record['go']['@source']
go_id_prop = source_ref_id[go_source['_id']]
reference = make_ref_source(go_source, go_id_prop,
external_ids[go_id_prop])
for go_level, go_records in record['go']['@value'].items():
level_wdid = go_props[go_level]
for go_record in go_records:
go_wdid = go_wdid_mapping[go_record['id']]
evidence_wdid = go_evidence_codes[go_record['evidence']]
evidence_statement = PBB_Core.WDItemID(value=evidence_wdid,
prop_nr='P459',
is_qualifier=True)
this_reference = copy.deepcopy(reference)
if add_pubmed:
for pubmed in go_record['pubmed']:
pmid_wdid = PBB_Helpers.PubmedStub(pubmed).create(login)
this_reference.append(
PBB_Core.WDItemID(pmid_wdid, 'P248',
is_reference=True))
s.append(
PBB_Core.WDItemID(go_wdid,
level_wdid,
references=[this_reference],
qualifiers=[evidence_statement]))
############
# statements with no referencable sources (make by hand, for now...)
############
# subclass of protein
s.append(PBB_Core.WDItemID('Q8054', 'P279', references=[ensembl_ref]))
# found in taxon
s.append(
PBB_Core.WDItemID(strain_info['organism_wdid'],
'P703',
references=[ensembl_ref]))
# encodes gene
s.append(PBB_Core.WDItemID(gene_qid, 'P702', references=[ensembl_ref]))
try:
wd_item_protein = PBB_Core.WDItemEngine(
item_name=item_name,
domain='proteins',
data=s,
append_value=['P279'],
fast_run=True,
fast_run_base_filter={
'P352': '',
'P703': strain_info['organism_wdid']
})
wd_item_protein.set_label(item_name)
wd_item_protein.set_description(item_description, lang='en')
wd_item_protein.set_aliases(
[record['symbol']['@value'], record['locus_tag']['@value']])
except Exception as e:
print(e)
PBB_Core.WDItemEngine.log(
"ERROR",
format_msg(record['entrezgene']['@value'], str(e), None,
ENTREZ_PROP))
return
try_write(wd_item_protein, record['entrezgene']['@value'], 'P351', login)
def create_protein_ipr(uniprot_id, uniprot_wdid, families, has_part,
release_info, login):
"""
Create interpro relationships to one protein
:param uniprot_id: uniprot ID of the protein to modify
:type uniprot_id: str
:param uniprot_wdid: wikidata ID of the protein
:param families: list of ipr wd ids the protein is a (P279) subclass of
:param has_part: list of ipr wd ids the protein has (P527) has part
:return:
"""
date = release_info['date']
version = release_info['version']
# create ref
ref_stated_in = PBB_Core.WDItemID("Q3047275", 'P248', is_reference=True)
ref_imported = PBB_Core.WDItemID("Q3047275", 'P143', is_reference=True)
ref_version = PBB_Core.WDString(version, 'P348', is_reference=True)
ref_date = PBB_Core.WDTime(date.strftime("+%Y-%m-%dT00:00:00Z"),
'P577',
is_reference=True)
ref_ipr = PBB_Core.WDString(
"http://www.ebi.ac.uk/interpro/protein/{}".format(uniprot_id),
"P854",
is_reference=True)
reference = [ref_stated_in, ref_imported, ref_version, ref_date, ref_ipr]
for ref in reference:
ref.overwrite_references = True
statements = []
if families:
for f in families:
statements.append(
PBB_Core.WDItemID(value=f,
prop_nr='P279',
references=[reference]))
if has_part:
for hp in has_part:
statements.append(
PBB_Core.WDItemID(value=hp,
prop_nr='P527',
references=[reference]))
item = PBB_Core.WDItemEngine(wd_item_id=uniprot_wdid,
data=statements,
server=SERVER,
append_value=["P279", "P527", "P361"])
# print(item.get_wd_json_representation())
try:
item.write(login)
except WDApiError as e:
print(e)
PBB_Core.WDItemEngine.log(
'ERROR',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'
.format(main_data_id=uniprot_id,
exception_type=type(e),
message=e.__str__(),
wd_id=uniprot_wdid,
duration=datetime.now()))
return
PBB_Core.WDItemEngine.log(
'INFO',
'{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'.
format(main_data_id=uniprot_id,
exception_type='',
message='created protein interpro relationships: {}'.format([
(x.prop_nr, x.value) for x in statements
]),
wd_id=uniprot_wdid,
duration=datetime.now()))
def make_chroms(strain_info, retrieved, login):
chrom_wdid = {}
for chrom_num, genome_id in strain_info['chrom_genomeid_map'].items():
item_name = '{} chromosome {}'.format(strain_info['organism_name'],
chrom_num)
item_description = '{} chromosome'.format(strain_info['organism_type'])
print(item_name)
print(genome_id)
reference = make_ref(retrieved, genome_id)
statements = []
statements.append(
PBB_Core.WDItemID(value='Q37748',
prop_nr='P279',
references=[reference
])) # subclass of chromosome
statements.append(
PBB_Core.WDItemID(value=strain_info['organism_wdid'],
prop_nr='P703',
references=[reference])) # found in taxon
statements.append(
PBB_Core.WDString(value=genome_id,
prop_nr='P2249',
references=[reference])) # genome id
wd_item = PBB_Core.WDItemEngine(item_name=item_name,
domain='chromosome',
data=statements,
append_value=['P279'],
fast_run=True,
fast_run_base_filter={
'P703':
strain_info['organism_wdid'],
'P2249': ''
})
if wd_item.require_write:
print("require write")
wd_item.set_label(item_name)
wd_item.set_description(item_description, lang='en')
try:
msg = "CREATE" if wd_item.create_new_item else "UPDATE"
wd_item.write(login=login)
PBB_Core.WDItemEngine.log(
"INFO",
format_msg(genome_id,
msg,
wd_item.wd_item_id,
external_id_prop='P2249'))
except Exception as e:
print(e)
PBB_Core.WDItemEngine.log(
"ERROR",
format_msg(genome_id,
str(e),
wd_item.wd_item_id,
external_id_prop='P2249'))
else:
chrom_wdid[chrom_num] = wd_item.wd_item_id
PBB_Core.WDItemEngine.log(
"INFO",
format_msg(genome_id,
"SKIP",
wd_item.wd_item_id,
external_id_prop='P2249'))
return chrom_wdid
def gene_item_statements():
"""
construct list of referenced statements to past to PBB_Core Item engine
"""
s = []
############
# external IDs
############
# will be used for reference statements
external_ids = {
'entrez_gene': str(record['entrezgene']['@value']),
'ensembl_gene': record['ensembl']['@value']['gene'],
'locus_tag': record['locus_tag']['@value']
}
# entrez gene id
entrez_ref = make_ref_source(record['entrezgene']['@source'],
'entrez_gene',
external_ids['entrez_gene'])
s.append(
PBB_Core.WDString(external_ids['entrez_gene'],
PROPS['Entrez Gene ID'],
references=[entrez_ref]))
# ensembl gene id
ensembl_ref = make_ref_source(record['ensembl']['@source'],
'ensembl_gene',
external_ids['ensembl_gene'])
s.append(
PBB_Core.WDString(external_ids['ensembl_gene'],
PROPS['Ensembl Gene ID'],
references=[ensembl_ref]))
# ncbi locus tag
s.append(
PBB_Core.WDString(external_ids['locus_tag'],
PROPS['NCBI Locus tag'],
references=[entrez_ref]))
############
# statements with no referencable sources (make by hand, for now...)
############
# subclass of gene
s.append(
PBB_Core.WDItemID('Q7187',
PROPS['subclass of'],
references=[ensembl_ref]))
# found in taxon
s.append(
PBB_Core.WDItemID(strain_info['organism_wdid'],
PROPS['found in taxon'],
references=[ensembl_ref]))
############
# genomic position: start, end, strand orientation, chromosome
############
genomic_pos_value = record['genomic_pos']['@value']
genomic_pos_source = record['genomic_pos']['@source']
genomic_pos_id_prop = source_ref_id[genomic_pos_source['_id']]
genomic_pos_ref = make_ref_source(genomic_pos_source,
genomic_pos_id_prop,
external_ids[genomic_pos_id_prop])
# create chromosome qualifier
chrom_genomeid = strain_info['chrom_genomeid_map'][
genomic_pos_value['chr']]
rs_chrom = PBB_Core.WDString(chrom_genomeid,
'P2249',
is_qualifier=True) # Refseq Genome ID
# strand orientation
strand_orientation = 'Q22809680' if genomic_pos_value[
'strand'] == 1 else 'Q22809711'
s.append(
PBB_Core.WDItemID(strand_orientation,
PROPS['strand orientation'],
references=[genomic_pos_ref]))
# genomic start and end
s.append(
PBB_Core.WDString(str(int(genomic_pos_value['start'])),
PROPS['genomic start'],
references=[genomic_pos_ref],
qualifiers=[rs_chrom]))
s.append(
PBB_Core.WDString(str(int(genomic_pos_value['end'])),
PROPS['genomic end'],
references=[genomic_pos_ref],
qualifiers=[rs_chrom]))
# chromosome
chr_genomic_id = strain_info['chrom_genomeid_map'][
genomic_pos_value['chr']]
s.append(
PBB_Core.WDItemID(chrom_wdid[chr_genomic_id],
PROPS['chromosome'],
references=[genomic_pos_ref]))
return s
def __init__(self, object):
"""
:type self: object
"""
self.start = object["start"]
self.entrezgene = object["entrezgene"]
self.uniprotwikidataids = object["uniprotwikidataids"]
gene_annotations = self.annotate_gene()
self.genomeInfo = object["speciesInfo"][str(gene_annotations['taxid'])]
self.content = object
self.name = gene_annotations["name"]
self.logincreds = object["logincreds"]
if "_timestamp" in gene_annotations.keys():
self.annotationstimestamp = gene_annotations["_timestamp"]
self.wdid = object["wdid"]
# symbol:
self.symbol = gene_annotations["symbol"]
print(self.symbol)
# HGNC
if "HGNC" in gene_annotations:
if isinstance(gene_annotations["HGNC"], list):
self.hgnc = gene_annotations["HGNC"]
else:
self.hgnc = [gene_annotations["HGNC"]]
else:
self.hgnc = None
# Ensembl Gene & transcript
if "ensembl" in gene_annotations:
if "gene" in gene_annotations["ensembl"]:
if isinstance(gene_annotations["ensembl"]["gene"], list):
self.ensembl_gene = gene_annotations["ensembl"]["gene"]
else:
self.ensembl_gene = [gene_annotations["ensembl"]["gene"]]
else:
self.ensembl_gene = None
if "transcript" in gene_annotations["ensembl"]:
if isinstance(gene_annotations["ensembl"]["transcript"], list):
self.ensembl_transcript = gene_annotations["ensembl"]["transcript"]
else:
self.ensembl_transcript = [gene_annotations["ensembl"]["transcript"]]
else:
self.ensembl_transcript = None
# Homologene
if "homologene" in gene_annotations:
if isinstance(gene_annotations["homologene"]["id"], list):
self.homologene = [str(i) for i in gene_annotations["homologene"]["id"]]
else:
self.homologene = [str(gene_annotations["homologene"]["id"])]
else:
self.homologene = None
# Refseq
if "refseq" in gene_annotations:
if "rna" in gene_annotations["refseq"]:
if isinstance(gene_annotations["refseq"]["rna"], list):
self.refseq_rna = gene_annotations["refseq"]["rna"]
else:
self.refseq_rna = [gene_annotations["refseq"]["rna"]]
else:
self.refseq_rna = None
else:
self.refseq_rna = None
# MGI
if "MGI" in gene_annotations:
if isinstance(gene_annotations["MGI"], list):
self.MGI = gene_annotations["MGI"]
else:
self.MGI = [gene_annotations["MGI"]]
else:
self.MGI = None
self.chromosome = None
self.startpost = None
self.endpos = None
if "genomic_pos" in gene_annotations:
if isinstance(gene_annotations["genomic_pos"], list):
self.chromosome = []
self.startpos = []
self.endpos = []
for i in range(len(gene_annotations["genomic_pos"])):
if gene_annotations["genomic_pos"][i]["chr"] in ProteinBoxBotKnowledge.chromosomes[
self.genomeInfo["name"]].keys():
self.chromosome.append(ProteinBoxBotKnowledge.chromosomes[self.genomeInfo["name"]][
gene_annotations["genomic_pos"][i]["chr"]])
self.startpos.append(gene_annotations["genomic_pos"][i]["start"])
self.endpos.append(gene_annotations["genomic_pos"][i]["end"])
else:
self.chromosome = []
self.startpos = []
self.endpos = []
if gene_annotations["genomic_pos"]["chr"] in ProteinBoxBotKnowledge.chromosomes[
self.genomeInfo["name"]].keys():
self.chromosome.append(ProteinBoxBotKnowledge.chromosomes[self.genomeInfo["name"]][
gene_annotations["genomic_pos"]["chr"]])
self.startpos.append(gene_annotations["genomic_pos"]["start"])
self.endpos.append(gene_annotations["genomic_pos"]["end"])
self.encodes = None
if "uniprot" in gene_annotations.keys():
if "Swiss-Prot" in gene_annotations["uniprot"].keys():
if isinstance(gene_annotations["uniprot"]["Swiss-Prot"], list):
self.encodes = []
for uniprot in gene_annotations["uniprot"]["Swiss-Prot"]:
self.encodes.append(uniprot)
else:
self.encodes = [gene_annotations["uniprot"]["Swiss-Prot"]]
self.chromosomeHg19 = None
self.startposHg19 = None
self.endposHg19 = None
if "genomic_pos_hg19" in gene_annotations:
if isinstance(gene_annotations["genomic_pos_hg19"], list):
self.chromosomeHg19 = []
self.startposHg19 = []
self.endposHg19 = []
for i in range(len(gene_annotations["genomic_pos_hg19"])):
if gene_annotations["genomic_pos_hg19"][i]["chr"] in ProteinBoxBotKnowledge.chromosomes[
self.genomeInfo["name"]].keys():
self.chromosomeHg19.append(ProteinBoxBotKnowledge.chromosomes[self.genomeInfo["name"]][
gene_annotations["genomic_pos_hg19"][i]["chr"]])
self.startposHg19.append(gene_annotations["genomic_pos_hg19"][i]["start"])
self.endposHg19.append(gene_annotations["genomic_pos_hg19"][i]["end"])
else:
self.chromosomeHg19 = []
self.startposHg19 = []
self.endposHg19 = []
if gene_annotations["genomic_pos_hg19"]["chr"] in ProteinBoxBotKnowledge.chromosomes[
self.genomeInfo["name"]].keys():
self.chromosomeHg19.append(ProteinBoxBotKnowledge.chromosomes[self.genomeInfo["name"]][
gene_annotations["genomic_pos_hg19"]["chr"]])
self.startposHg19.append(gene_annotations["genomic_pos_hg19"]["start"])
self.endposHg19.append(gene_annotations["genomic_pos_hg19"]["end"])
# type of Gene
if "type_of_gene" in gene_annotations:
self.type_of_gene = []
if gene_annotations["type_of_gene"] == "ncRNA":
self.type_of_gene.append("Q427087")
if gene_annotations["type_of_gene"] == "snRNA":
self.type_of_gene.append("Q284578")
if gene_annotations["type_of_gene"] == "snoRNA":
self.type_of_gene.append("Q284416")
if gene_annotations["type_of_gene"] == "rRNA":
self.type_of_gene.append("Q215980")
if gene_annotations["type_of_gene"] == "tRNA":
self.type_of_gene.append("Q201448")
if gene_annotations["type_of_gene"] == "pseudo":
self.type_of_gene.append("Q277338")
if gene_annotations["type_of_gene"] == "protein-coding":
self.type_of_gene.append("Q20747295")
else:
self.type_of_gene = None
# Reference section
# Prepare references
refStatedIn = PBB_Core.WDItemID(value=self.genomeInfo["release"], prop_nr='P248', is_reference=True)
refStatedIn.overwrite_references = True
refImported = PBB_Core.WDItemID(value='Q20641742', prop_nr='P143', is_reference=True)
refImported.overwrite_references = True
timeStringNow = strftime("+%Y-%m-%dT00:00:00Z", gmtime())
refRetrieved = PBB_Core.WDTime(timeStringNow, prop_nr='P813', is_reference=True)
refRetrieved.overwrite_references = True
gene_reference = [refStatedIn, refImported, refRetrieved]
refStatedInEnsembl = PBB_Core.WDItemID(value= 'Q21996330', prop_nr='P248', is_reference=True)
refStatedInEnsembl.overwrite_references = True
refImportedEnsembl = PBB_Core.WDItemID(value='Q1344256', prop_nr='P143', is_reference=True)
refImportedEnsembl.overwrite_references = True
ensembl_reference = [refStatedInEnsembl, refImportedEnsembl, refRetrieved]
genomeBuildQualifier = PBB_Core.WDItemID(value=self.genomeInfo["genome_assembly"], prop_nr='P659',
is_qualifier=True)
genomeBuildPreviousQualifier = PBB_Core.WDItemID(value=self.genomeInfo["genome_assembly_previous"],
prop_nr='P659', is_qualifier=True)
prep = dict()
prep['P703'] = [PBB_Core.WDItemID(value=self.genomeInfo['wdid'], prop_nr='P703',
references=[copy.deepcopy(gene_reference)])]
if self.genomeInfo["name"] == "human":
prep['P353'] = [
PBB_Core.WDString(value=self.symbol, prop_nr='P353', references=[copy.deepcopy(gene_reference)])]
prep['P351'] = [
PBB_Core.WDString(value=str(self.entrezgene), prop_nr='P351', references=[copy.deepcopy(gene_reference)])]
prep['P279'] = [PBB_Core.WDItemID(value='Q7187', prop_nr='P279', references=[copy.deepcopy(gene_reference)])]
if "type_of_gene" in vars(self):
if self.type_of_gene != None:
for i in range(len(self.type_of_gene)):
prep['P279'].append(PBB_Core.WDItemID(value=self.type_of_gene[i], prop_nr='P279',
references=[copy.deepcopy(gene_reference)]))
if "ensembl_gene" in vars(self):
if self.ensembl_gene != None:
prep['P594'] = []
for ensemblg in self.ensembl_gene:
prep['P594'].append(
PBB_Core.WDString(value=ensemblg, prop_nr='P594', references=[copy.deepcopy(gene_reference)]))
if "ensembl_transcript" in vars(self):
if self.ensembl_transcript != None:
prep['P704'] = []
for ensemblt in self.ensembl_transcript:
prep['P704'].append(
PBB_Core.WDString(value=ensemblt, prop_nr='P704', references=[copy.deepcopy(gene_reference)]))
if "encodes" in vars(self):
if self.encodes != None:
prep['P688'] = []
for uniprot in self.encodes:
if uniprot in self.uniprotwikidataids.keys():
prep['P688'].append(PBB_Core.WDItemID(value=self.uniprotwikidataids[uniprot], prop_nr='P688', references=[copy.deepcopy(gene_reference)]))
if "hgnc" in vars(self):
if self.hgnc != None:
prep['P354'] = []
for hugo in self.hgnc:
prep['P354'].append(
PBB_Core.WDString(value=hugo, prop_nr='P354', references=[copy.deepcopy(gene_reference)]))
if "homologene" in vars(self):
if self.homologene != None:
prep['P593'] = []
for ortholog in self.homologene:
prep['P593'].append(
PBB_Core.WDString(value=ortholog, prop_nr='P593', references=[copy.deepcopy(gene_reference)]))
if "refseq_rna" in vars(self):
if self.refseq_rna != None:
prep['P639'] = []
for refseq in self.refseq_rna:
prep['P639'].append(
PBB_Core.WDString(value=refseq, prop_nr='P639', references=[copy.deepcopy(gene_reference)]))
if "chromosome" in vars(self):
prep['P1057'] = []
if self.chromosome != None:
for chrom in list(set(self.chromosome)):
prep['P1057'].append(
PBB_Core.WDItemID(value=chrom, prop_nr='P1057', references=[copy.deepcopy(gene_reference)]))
if "startpos" in vars(self):
if not 'P644' in prep.keys():
prep['P644'] = []
if self.startpos != None:
for pos in self.startpos:
prep['P644'].append(
PBB_Core.WDString(value=str(pos), prop_nr='P644', references=[copy.deepcopy(ensembl_reference)],
qualifiers=[copy.deepcopy(genomeBuildQualifier)]))
if "endpos" in vars(self):
if not 'P645' in prep.keys():
prep['P645'] = []
if self.endpos != None:
for pos in self.endpos:
prep['P645'].append(
PBB_Core.WDString(value=str(pos), prop_nr='P645', references=[copy.deepcopy(ensembl_reference)],
qualifiers=[copy.deepcopy(genomeBuildQualifier)]))
if "startposHg19" in vars(self):
if not 'P644' in prep.keys():
prep['P644'] = []
if self.startposHg19 != None:
for pos in self.startposHg19:
prep['P644'].append(
PBB_Core.WDString(value=str(pos), prop_nr='P644', references=[copy.deepcopy(ensembl_reference)],
qualifiers=[copy.deepcopy(genomeBuildPreviousQualifier)]))
if "endposHg19" in vars(self):
if not 'P644' in prep.keys():
prep['P645'] = []
if self.endposHg19 != None:
for pos in self.endposHg19:
prep['P645'].append(
PBB_Core.WDString(value=str(pos), prop_nr='P645', references=[copy.deepcopy(ensembl_reference)],
qualifiers=[copy.deepcopy(genomeBuildPreviousQualifier)]))
if "MGI" in vars(self):
prep['P671'] = []
if self.MGI != None:
for mgi in self.MGI:
prep['P671'].append(PBB_Core.WDString(value=mgi, prop_nr='P671',
references=[copy.deepcopy(gene_reference)]))
if "alias" in gene_annotations.keys():
if isinstance(gene_annotations["alias"], list):
self.synonyms = []
for alias in gene_annotations["alias"]:
self.synonyms.append(alias)
else:
self.synonyms = [gene_annotations["alias"]]
self.synonyms.append(self.symbol)
print(self.synonyms)
else:
self.synonyms = None
data2add = []
for key in prep.keys():
for statement in prep[key]:
data2add.append(statement)
print(statement.prop_nr, statement.value)
if self.wdid != None:
# if self.encodes != None:
wdPage = PBB_Core.WDItemEngine(self.wdid, item_name=self.name, data=data2add, server="www.wikidata.org",
domain="genes")
if wdPage.get_description() == "":
wdPage.set_description(description=self.genomeInfo['name'] + ' gene', lang='en')
if wdPage.get_description(lang='fr') == "" or wdPage.get_description(lang='fr') == "gène":
wdPage.set_description(description="Un gène " + self.genomeInfo['fr-name'], lang='fr')
if wdPage.get_description(lang='nl') == "" or wdPage.get_description(lang='nl') == "gen":
wdPage.set_description(description="Een "+ self.genomeInfo['nl-name']+ " gen", lang='nl')
if self.synonyms != None:
wdPage.set_aliases(aliases=self.synonyms, lang='en', append=True)
print(self.wdid)
self.wd_json_representation = wdPage.get_wd_json_representation()
PBB_Debug.prettyPrint(self.wd_json_representation)
PBB_Debug.prettyPrint(data2add)
# print(self.wd_json_representation)
wdPage.write(self.logincreds)
print("aa")
else:
#if self.encodes != None:
wdPage = PBB_Core.WDItemEngine(item_name=self.name, data=data2add, server="www.wikidata.org",
domain="genes")
if wdPage.get_description() != "":
wdPage.set_description(description=self.genomeInfo['name'] + ' gene', lang='en')
if wdPage.get_description(lang='fr') == "" or wdPage.get_description(lang='fr') == "gène":
wdPage.setdescription(description="Un gène " + self.genomeInfo['fr-name'], lang='fr')
if wdPage.get_description(lang='nl') == "" or wdPage.get_description(lang='nl') == "gen":
wdPage.setdescription(description="Een "+ self.genomeInfo['nl-name']+ " gen", lang='nl')
if self.synonyms != None:
wdPage.set_aliases(aliases=self.synonyms, lang='en', append=True)
self.wd_json_representation = wdPage.get_wd_json_representation()
PBB_Debug.prettyPrint(self.wd_json_representation)
PBB_Debug.prettyPrint(data2add)
# print(self.wd_json_representation)
self.wdid = wdPage.write(self.logincreds)
PBB_Core.WDItemEngine.log('INFO', '{main_data_id}, "{exception_type}", "{message}", {wd_id}, {duration}'.format(
main_data_id=str(self.entrezgene),
exception_type='',
message=f.name,
wd_id=self.wdid,
duration=time.time()-self.start
))