class HGNCParser(ReturnParser):
def __init__(self):
"""
:param ncbigene_instance: NcbiGene Instance
:type ncbigene_instance: DataSourceInstance
:param taxid:
"""
super(HGNCParser, self).__init__()
# output data
self.genes = NodeSet(['Gene'], merge_keys=['sid'])
self.gene_maps_gene = RelationshipSet('MAPS', ['Gene'], ['Gene'],
['sid'], ['sid'])
self.gene_maps_genesymbol = RelationshipSet('MAPS', ['Gene'],
['GeneSymbol'], ['sid'],
['sid', 'taxid'])
def run_with_mounted_arguments(self):
self.run()
def run(self):
hgnc_instance = self.get_instance_by_name('HGNC')
hgnc_complete_file = hgnc_instance.get_file('hgnc_complete_set.txt')
self.parse_hgnc_complete_file(hgnc_complete_file)
def parse_hgnc_complete_file(self, hgnc_complete_file):
with open(hgnc_complete_file, 'rt') as f:
header = next(f)
for l in f:
flds = l.strip().split('\t')
sid = flds[0]
gene_symbol = flds[1]
ncbi_id = flds[18] if len(flds) > 18 else None
ensembl_id = flds[19] if len(flds) > 19 else None
all_props = dict(zip(header, flds))
all_props['sid'] = sid
all_props['source'] = 'hgnc'
self.genes.add_node(all_props)
if ncbi_id:
self.gene_maps_gene.add_relationship({'sid': sid},
{'sid': ncbi_id},
{'source': 'hgnc'})
if ensembl_id:
self.gene_maps_gene.add_relationship({'sid': sid},
{'sid': ensembl_id},
{'source': 'hgnc'})
if gene_symbol:
self.gene_maps_genesymbol.add_relationship({'sid': sid}, {
'sid': gene_symbol,
'taxid': '9606'
}, {'source': 'hgnc'})
def __init__(self):
super(SomeParser, self).__init__()
self.source = NodeSet(['Source'], merge_keys=['source_id'])
self.target = NodeSet(['Target'], merge_keys=['target_id'])
self.rels = RelationshipSet('FOO', ['Source'], ['Target'],
['source_id'], ['target_id'])
def __init__(self):
super(NcbiGeneParser, self).__init__()
# arguments
self.arguments = ['taxid']
# output data
# both gene IDs and GeneSymbols have the label 'Gene'
# two different NodeSets are used because only the GeneSymbol nodes need taxid for uniqueness
self.genes = NodeSet(['Gene'],
merge_keys=['sid'],
default_props={'source': 'ncbigene'})
self.genesymbols = NodeSet(['Gene'],
merge_keys=['sid', 'taxid'],
default_props={
'source': 'ncbigene',
'type': 'symbol'
})
self.genesymbol_synonym_genesymbol = RelationshipSet(
'SYNONYM', ['Gene'], ['Gene'], ['sid', 'taxid'], ['sid', 'taxid'],
default_props={'source': 'ncbigene'})
self.gene_maps_genesymbol = RelationshipSet(
'MAPS', ['Gene'], ['Gene'], ['sid'], ['sid', 'taxid'],
default_props={'source': 'ncbigene'})
def __init__(self):
"""
"""
super(BigWordListParser, self).__init__()
# NodeSets
self.words = NodeSet(['Word'], merge_keys=['value'])
class DummyParser(ReturnParser):
def __init__(self):
super(DummyParser, self).__init__()
# arguments
self.arguments = ['taxid']
# output data
self.dummy_nodes = NodeSet(['Dummy'], merge_keys=['sid'])
self.fummy_nodes = NodeSet(['Fummy'], merge_keys=['sid'])
self.dummy_knows_fummy = RelationshipSet('KNOWS', ['Dummy'], ['Fummy'],
['sid'], ['sid'])
def run_with_mounted_arguments(self):
self.run(self.taxid)
def run(self, taxid):
dummy_instance = self.get_instance_by_name('Dummy')
dummyfile = dummy_instance.get_file('file.txt')
target_sids = list(string.ascii_lowercase)
# Fummy nodes
for i in range(10):
self.fummy_nodes.add_node({'sid': i, 'taxid': taxid})
with open(dummyfile) as f:
for l in f:
letter = l.strip()
self.dummy_nodes.add_node({'sid': letter, 'taxid': taxid})
self.dummy_knows_fummy.add_relationship({'sid': letter},
{'sid': randint(0, 9)},
{'key': 'value'})
def __init__(self):
super(LncipediaParser, self).__init__()
self.genes = NodeSet(['Gene'], merge_keys=['sid'])
self.transcripts = NodeSet(['Transcript'], merge_keys=['sid'])
self.gene_codes_transcripts = RelationshipSet('CODES', ['Gene'], ['Transcript'], ['sid'], ['sid'])
self.gene_maps_gene = RelationshipSet('MAPS', ['Gene'], ['Gene'], ['sid'], ['sid'])
self.transcript_maps_transcript = RelationshipSet('MAPS', ['Transcript'], ['Transcript'], ['sid'], ['sid'])
def __init__(self):
super(RefseqEntityParser, self).__init__()
# arguments
self.arguments = ['taxid']
# define NodeSet and RelationshipSet
self.transcripts = NodeSet(['Transcript'], merge_keys=['sid'], default_props={'source': 'refseq'})
self.proteins = NodeSet(['Protein'], merge_keys=['sid'], default_props={'source': 'refseq'})
def __init__(self):
super(NcbiLegacyGeneParser, self).__init__()
self.arguments = ['taxid']
self.legacy_genes = NodeSet(['Gene', 'Legacy'],
merge_keys=['sid'],
default_props={'source': 'ncbigene'})
self.legacy_gene_now_gene = RelationshipSet(
'REPLACED_BY', ['Gene', 'Legacy'], ['Gene'], ['sid'], ['sid'],
default_props={'source': 'ncbigene'})
def __init__(self):
super(DummyParser, self).__init__()
# arguments
self.arguments = ['taxid']
# output data
self.dummy_nodes = NodeSet(['Dummy'], merge_keys=['sid'])
self.fummy_nodes = NodeSet(['Fummy'], merge_keys=['sid'])
self.dummy_knows_fummy = RelationshipSet('KNOWS', ['Dummy'], ['Fummy'],
['sid'], ['sid'])
def __init__(self):
super(SwissLipidsParser, self).__init__()
# define NodeSet and RelationshipSet
self.lipids = NodeSet(['Lipid'], merge_keys=['sid'])
self.lipid_fromclass_lipid = RelationshipSet('FROM_LIPID_CLASS', ['Lipid'], ['Lipid'], ['sid'], ['sid'])
self.lipid_parent_lipid = RelationshipSet('HAS_PARENT', ['Lipid'], ['Lipid'], ['sid'], ['sid'])
self.lipid_component_lipid = RelationshipSet('HAS_COMPONENT', ['Lipid'], ['Lipid'], ['sid'], ['sid'])
self.lipid_maps_metabolite = RelationshipSet('MAPS', ['Lipid'], ['Metabolite'], ['sid'], ['sid'])
self.lipid_associates_protein = RelationshipSet('HAS_ASSOCIATION', ['Lipid'], ['Protein'], ['sid'], ['sid'])
def __init__(self):
"""
:param ensembl_instance: The ENSEMBL DataSource instance.
"""
super(EnsemblLocusParser, self).__init__()
# arguments
self.arguments = ['taxid']
# NodeSets
self.locus = NodeSet(['Locus'],
merge_keys=['uuid'],
default_props={'source': 'ensembl'})
class RootTestParser(ReturnParser):
def __init__(self):
super(RootTestParser, self).__init__()
self.source = NodeSet(['Source'], merge_keys=['source_id'])
self.target = NodeSet(['Target'], merge_keys=['target_id'])
def run_with_mounted_arguments(self):
self.run()
def run(self):
for i in range(100):
self.source.add_node({'source_id': i})
self.target.add_node({'target_id': i})
class EnsemblLocusParser(ReturnParser):
def __init__(self):
"""
:param ensembl_instance: The ENSEMBL DataSource instance.
"""
super(EnsemblLocusParser, self).__init__()
# arguments
self.arguments = ['taxid']
# NodeSets
self.locus = NodeSet(['Locus'],
merge_keys=['uuid'],
default_props={'source': 'ensembl'})
def run_with_mounted_arguments(self):
self.run(self.taxid)
def run(self, taxid):
ensembl_instance = self.get_instance_by_name('Ensembl')
datasource_name = ensembl_instance.datasource.name
ensembl_gtf_file_path = Ensembl.get_gtf_file_path(
taxid, ensembl_instance)
annotation = GffReader(ensembl_gtf_file_path)
log.info("Start parsing ENSEMBL gtf file, taxid {}, {}".format(
taxid, ensembl_gtf_file_path))
for r in annotation.records:
# one line is one unique Locus
props = {
'chr': r.chr,
'annotation_source': r.source,
'start': int(r.start),
'end': int(r.end),
'type': r.type,
'score': r.score,
'strand': r.strand,
'frame': r.frame,
'taxid': taxid,
'ref': 'h38',
'uuid': str(uuid4())
}
props.update(r.attributes)
self.locus.add_node(props)
log.info("Finished parsing ENSEMBL gtf file.")
def __init__(self):
super(HmdbParser, self).__init__()
# NodeSets
self.metabolites = NodeSet(['Metabolite'],
merge_keys=['sid'],
default_props={'source': 'hmdb'})
self.metabolite_map_metabolite = RelationshipSet(
'MAPS', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'hmdb'})
self.metabolite_associates_protein = RelationshipSet(
'HAS_ASSOCIATION', ['Metabolite'], ['Protein'], ['sid'], ['sid'],
default_props={'source': 'hmdb'})
class NcbiLegacyGeneParser(ReturnParser):
"""
Parse legacy gene IDs from gene_history.gz
#tax_id GeneID Discontinued_GeneID Discontinued_Symbol Discontinue_Date
9 - 1246494 repA1 20031113
9 - 1246495 repA2 20031113
9 - 1246496 leuA 20031113
"""
def __init__(self):
super(NcbiLegacyGeneParser, self).__init__()
self.arguments = ['taxid']
self.legacy_genes = NodeSet(['Gene', 'Legacy'],
merge_keys=['sid'],
default_props={'source': 'ncbigene'})
self.legacy_gene_now_gene = RelationshipSet(
'REPLACED_BY', ['Gene', 'Legacy'], ['Gene'], ['sid'], ['sid'],
default_props={'source': 'ncbigene'})
def run_with_mounted_arguments(self):
self.run(self.taxid)
def run(self, taxid):
log.debug(f'Run parser {self.__class__.__name__} for taxID: {taxid}.')
ncbigene_instance = self.get_instance_by_name('NcbiGene')
gene_history_file = ncbigene_instance.get_file('gene_history.gz')
with gzip.open(gene_history_file, 'rt') as f:
# skip header
next(f)
for l in f:
flds = l.strip().split('\t')
this_taxid = flds[0]
if this_taxid == taxid:
new_gene_id = flds[1]
discontinued_gene_id = flds[2]
discontinued_symbol = flds[3]
date = flds[4]
self.legacy_genes.add_node({
'sid': discontinued_gene_id,
'date': date,
'symbol': discontinued_symbol,
'taxid': taxid
})
if new_gene_id != '-':
self.legacy_gene_now_gene.add_relationship(
{'sid': discontinued_gene_id},
{'sid': new_gene_id}, {})
def __init__(self):
super(RefseqRemovedRecordsParser, self).__init__()
self.arguments = ['taxid']
self.legacy_ids = set()
self.legacy_transcripts = NodeSet(['Transcript', 'Legacy'], merge_keys=['sid'], default_props={'source': 'refseq'})
self.legacy_transcript_now_transcript = RelationshipSet('REPLACED_BY', ['Transcript'], ['Transcript'], ['sid'], ['sid'], default_props={'source': 'refseq'})
self.legacy_proteins = NodeSet(['Protein', 'Legacy'], merge_keys=['sid'], default_props={'source': 'refseq'})
self.legacy_protein_now_protein = RelationshipSet('REPLACED_BY', ['Protein'], ['Protein'],
['sid'], ['sid'], default_props={'source': 'refseq'})
self.gene_codes_legacy_transcript = RelationshipSet('CODES', ['Gene'], ['Transcript', 'Legacy'], ['sid'], ['sid'], default_props={'source': 'refseq'})
self.legacy_transcript_codes_protein = RelationshipSet('CODES', ['Transcript', 'Legacy'], ['Protein'],
['sid'], ['sid'], default_props={'source': 'refseq'})
def __init__(self):
super(ChebiParser, self).__init__()
# NodeSets
self.metabolites = NodeSet(['Metabolite'],
merge_keys=['sid'],
default_props={'source': 'chebi'})
self.metabolite_isa_metabolite = RelationshipSet(
'IS_A', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'chebi'})
self.metabolite_rel_metabolite = RelationshipSet(
'CHEBI_REL', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'chebi'})
self.metabolite_maps_metabolite = RelationshipSet(
'MAPS', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'chebi'})
def __init__(self):
super(MirbaseParser, self).__init__()
# NodeSets
self.precursor_mirna = NodeSet(['PrecursorMirna'], merge_keys=['sid'])
self.mature_mirna = NodeSet(['Mirna'], merge_keys=['sid'])
# RelationshipSets
self.precursor_codes_mature = RelationshipSet('PRE',
['PrecursorMirna'],
['Mirna'], ['sid'],
['sid'])
self.transcript_codes_precursor = RelationshipSet(
'IS', ['Transcript'], ['PrecursorMirna'], ['sid'], ['sid'])
self.gene_is_precursor = RelationshipSet('IS', ['Gene'],
['PrecursorMirna'], ['sid'],
['sid'])
def deserialize(cls, source_dir: str, metadata_only: bool = False) -> 'Parser':
"""
Read from a serialized directory, recreate a Parser that can load to the database.
:param source_dir: Directory to read from.
:return: A Parser object.
"""
log.debug(f"Read Parser from {source_dir}.")
p = cls()
for file in os.listdir(source_dir):
if not metadata_only:
if file.startswith('nodeset_'):
ns_name = file.replace('.json', '')
with open(os.path.join(source_dir, file), 'rt') as f:
log.debug(f"Deserialize {f}")
ns = NodeSet.from_dict(json.load(f))
log.debug(f"Num nodes in NodeSet: {len(ns.nodes)}")
p.__dict__[ns_name] = ns
elif file.startswith('relationshipset_'):
rs_name = file.replace('.json', '')
with open(os.path.join(source_dir, file), 'rt') as f:
log.debug(f"Deserialize {f}")
rs = RelationshipSet.from_dict(json.load(f))
log.debug(f"Num relationships in RelationshipSet: {len(rs.relationships)}")
p.__dict__[rs_name] = rs
if file == 'parser_data.json':
with open(os.path.join(source_dir, file), 'rt') as f:
metadata = json.load(f)
# TODO add datasource instances to deserializer
p.name = metadata['name']
return p
def __init__(self):
"""
:param uniprot_instance: The Uniprot instance
:param taxid: The taxid
"""
super(UniprotKnowledgebaseParser, self).__init__()
# arguments
self.arguments = ['taxid']
# NodeSet
self.proteins = NodeSet(['Protein'], merge_keys=['sid'], default_props={'source': 'uniprot'})
# RelationshipSet
self.protein_primary_protein = RelationshipSet('PRIMARY', ['Protein'], ['Protein'], ['sid'], ['sid'], default_props={'source': 'uniprot'})
self.transcript_codes_protein = RelationshipSet('CODES', ['Transcript'], ['Protein'], ['sid'], ['sid'], default_props={'source': 'uniprot'})
self.protein_maps_protein = RelationshipSet('MAPS', ['Protein'], ['Protein'], ['sid'], ['sid'], default_props={'source': 'uniprot'})
def __init__(self):
"""
:param ncbigene_instance: NcbiGene Instance
:type ncbigene_instance: DataSourceInstance
:param taxid:
"""
super(HGNCParser, self).__init__()
# output data
self.genes = NodeSet(['Gene'], merge_keys=['sid'])
self.gene_maps_gene = RelationshipSet('MAPS', ['Gene'], ['Gene'],
['sid'], ['sid'])
self.gene_maps_genesymbol = RelationshipSet('MAPS', ['Gene'],
['GeneSymbol'], ['sid'],
['sid', 'taxid'])
class SomeParser(ReturnParser):
def __init__(self):
super(SomeParser, self).__init__()
self.source = NodeSet(['Source'], merge_keys=['source_id'])
self.target = NodeSet(['Target'], merge_keys=['target_id'])
self.rels = RelationshipSet('FOO', ['Source'], ['Target'],
['source_id'], ['target_id'])
def run_with_mounted_arguments(self):
self.run()
def run(self):
for i in range(100):
self.source.add_node({'source_id': i})
self.target.add_node({'target_id': i})
self.rels.add_relationship({'source_id': i}, {'target_id': i},
{'source': 'test'})
def _add_node(self, node):
# create nodeSet if necessary
labels = frozenset(node.labels)
if not labels in self.nodeSets:
# get primary keys
self.nodeSets[labels] = NodeSet(
list(labels), merge_keys=self._get_merge_keys(node))
# add node to nodeset
self.nodeSets[labels].nodes.append(node)
class BigWordListParser(ReturnParser):
def __init__(self):
"""
"""
super(BigWordListParser, self).__init__()
# NodeSets
self.words = NodeSet(['Word'], merge_keys=['value'])
def run_with_mounted_arguments(self):
self.run()
def run(self):
log.info("Run {}".format(self.__class__.__name__))
bigwordlist_instance = self.get_instance_by_name('BigWordList')
# collect the words and a list of wordlists they are metioned in
word_to_list = {}
log.info("")
for i in range(3, 13):
match_file = bigwordlist_instance.get_file(
'wlist_match{}.txt'.format(i))
try:
log.info("Open {}".format(match_file))
with open(match_file, 'r') as f:
for l in f:
word = l.strip()
if word in word_to_list:
word_to_list[word].append(i)
else:
word_to_list[word] = [i]
except TypeError:
log.info("Cannot open file {}".format(match_file))
for word, list_of_matches in word_to_list.items():
node_props = {'value': word}
for i in list_of_matches:
node_props["match{}".format(i)] = True
self.words.add_node(node_props)
def __init__(self):
"""
:param ensembl_instance: The ENSEMBL DataSource instance.
"""
super(EnsemblEntityParser, self).__init__()
# arguments
self.arguments = ['taxid']
# NodeSets
self.genes = NodeSet(['Gene'],
merge_keys=['sid'],
default_props={'source': 'ensembl'})
self.transcripts = NodeSet(['Transcript'],
merge_keys=['sid'],
default_props={'source': 'ensembl'})
self.proteins = NodeSet(['Protein'],
merge_keys=['sid'],
default_props={'source': 'ensembl'})
# RelationshipSets
self.gene_codes_transcript = RelationshipSet(
'CODES', ['Gene'], ['Transcript'], ['sid'], ['sid'],
default_props={'source': 'ensembl'})
self.transcript_codes_protein = RelationshipSet(
'CODES', ['Transcript'], ['Protein'], ['sid'], ['sid'],
default_props={'source': 'ensembl'})
def __init__(self):
"""
:param mesh_instance: NcbiGene Instance
:type mesh_instance: DataSourceInstance
"""
super(GtexMetadataParser, self).__init__()
# NodeSets
self.tissues = NodeSet(['GtexTissue'], merge_keys=['name'])
self.detailed_tissues = NodeSet(['GtexDetailedTissue'],
merge_keys=['name'])
self.sample = NodeSet(['GtexSample'], merge_keys=['sid'])
self.sample_measures_tissue = RelationshipSet('MEASURES',
['GtexSample'],
['GtexTissue'], ['sid'],
['name'])
self.sample_measures_detailed_tissue = RelationshipSet(
'MEASURES', ['GtexSample'], ['GtexDetailedTissue'], ['sid'],
['name'])
self.tissue_parent_detailed_tissue = RelationshipSet(
'PARENT', ['GtexTissue'], ['GtexDetailedTissue'], ['name'],
['name'])
self.tissue_parent_detailed_tissue.unique = True
def __init__(self):
super(MeshParser, self).__init__()
# NodeSets
self.descriptor = NodeSet(['MeshDescriptor'], merge_keys=['sid'])
self.qualifier = NodeSet(['MeshQualifier'], merge_keys=['sid'])
self.concept = NodeSet(['MeshConcept'], merge_keys=['sid'])
self.term = NodeSet(['MeshTerm'], merge_keys=['sid'])
self.descriptor_allowed_qualifier = RelationshipSet('ALLOWED', ['MeshDescriptor'], ['MeshQualifier'], ['sid'],
['sid'])
self.descriptor_has_concept = RelationshipSet('HAS', ['MeshDescriptor'], ['MeshConcept'], ['sid'], ['sid'])
self.descriptor_has_concept.unique = True
self.concept_has_term = RelationshipSet('HAS', ['MeshConcept'], ['MeshTerm'], ['sid'], ['sid'])
self.concept_has_term.unique = True
self.concept_related_concept = RelationshipSet('RELATED', ['MeshConcept'], ['MeshConcept'], ['sid'], ['sid'])
self.concept_related_concept.unique = True
class ChebiParser(ReturnParser):
def __init__(self):
super(ChebiParser, self).__init__()
# NodeSets
self.metabolites = NodeSet(['Metabolite'],
merge_keys=['sid'],
default_props={'source': 'chebi'})
self.metabolite_isa_metabolite = RelationshipSet(
'IS_A', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'chebi'})
self.metabolite_rel_metabolite = RelationshipSet(
'CHEBI_REL', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'chebi'})
self.metabolite_maps_metabolite = RelationshipSet(
'MAPS', ['Metabolite'], ['Metabolite'], ['sid'], ['sid'],
default_props={'source': 'chebi'})
def run_with_mounted_arguments(self):
self.run()
def run(self):
chebi_instance = self.get_instance_by_name('Chebi')
obo_file = chebi_instance.get_file('chebi.obo')
cleaned_obo_file = clean_obo_file(obo_file)
chebi_ontology = pronto.Ontology(cleaned_obo_file)
reltypes = set()
# iterate terms
for term in chebi_ontology.terms():
term_sid = (term.id).split(':')[1]
ontology_id = term.id
self.metabolites.add_node({
'name': (term.name),
'sid': term_sid,
'ontology_id': ontology_id,
'definition': term.definition,
'alt_ids': list(term.alternate_ids)
})
for parent in term.superclasses(distance=1, with_self=False):
self.metabolite_isa_metabolite.add_relationship(
{'sid': term_sid}, {'sid': parent.id}, {})
## other named relationships
try:
for reltype, targets in term.relationships.items():
for target in targets:
self.metabolite_rel_metabolite.add_relationship(
{'sid': term_sid}, {'sid': target.id},
{'type': reltype.id})
except KeyError as e:
log.error(f"Cannot iterate relationshis of term {term_sid}")
log.error(e)
# metabolite-MAPS-metabolite
for xref in term.xrefs:
if 'HMDB:' in xref.id:
hmdb_id = xref.id.strip().split('HMDB:')[1]
self.metabolite_maps_metabolite.add_relationship(
{'sid': term_sid}, {'sid': hmdb_id}, {})
def load_wpp_data(base_path, graph):
"""
Load UN population data.
:param base_path: Path where file was downloaded.
"""
un_wpp_csv_file = os.path.join(base_path, 'WPP2019_PopulationByAgeSex_Medium.csv')
log.info('Parse UN population data file: {}'.format(un_wpp_csv_file))
country = NodeSet(['Country'], ['name'])
age_group_nodes = NodeSet(['AgeGroup'], ['group'])
country_total_group = RelationshipSet('CURRENT_TOTAL', ['Country'], ['AgeGroup'], ['name'], ['group'])
country_male_group = RelationshipSet('CURRENT_MALE', ['Country'], ['AgeGroup'], ['name'], ['group'])
country_female_group = RelationshipSet('CURRENT_FEMALE', ['Country'], ['AgeGroup'], ['name'], ['group'])
countries_added = set()
age_groups_added = set()
with open(un_wpp_csv_file, 'rt') as f:
csv_file = csv.reader(f, delimiter=',', quotechar='"')
# skip header
next(csv_file)
for row in csv_file:
# LocID,Location,VarID,Variant,Time,MidPeriod,AgeGrp,AgeGrpStart,AgeGrpSpan,PopMale,PopFemale,PopTotal
loc_id = row[0]
location = row[1]
time = int(row[4])
age_group = row[6]
age_group_start = int(row[7])
age_group_span = row[8]
pop_male = int(float((row[9])) * 1000)
pop_female = int(float((row[10])) * 1000)
pop_total = int(float((row[11])) * 1000)
# only take 2019
if time == 2019:
if location not in countries_added:
country.add_node({'name': location, 'un_id': loc_id})
countries_added.add(location)
if age_group not in age_groups_added:
age_group_nodes.add_node({'group': age_group, 'start': age_group_start, 'span': age_group_span})
country_total_group.add_relationship({'name': location}, {'group': age_group}, {'count': pop_total})
country_male_group.add_relationship({'name': location}, {'group': age_group}, {'count': pop_male})
country_female_group.add_relationship({'name': location}, {'group': age_group}, {'count': pop_female})
log.info('Load data to Neo4j')
country.merge(graph)
age_group_nodes.merge(graph)
country_total_group.merge(graph)
country_male_group.merge(graph)
country_female_group.merge(graph)
class EnsemblEntityParser(ReturnParser):
def __init__(self):
"""
:param ensembl_instance: The ENSEMBL DataSource instance.
"""
super(EnsemblEntityParser, self).__init__()
# arguments
self.arguments = ['taxid']
# NodeSets
self.genes = NodeSet(['Gene'],
merge_keys=['sid'],
default_props={'source': 'ensembl'})
self.transcripts = NodeSet(['Transcript'],
merge_keys=['sid'],
default_props={'source': 'ensembl'})
self.proteins = NodeSet(['Protein'],
merge_keys=['sid'],
default_props={'source': 'ensembl'})
# RelationshipSets
self.gene_codes_transcript = RelationshipSet(
'CODES', ['Gene'], ['Transcript'], ['sid'], ['sid'],
default_props={'source': 'ensembl'})
self.transcript_codes_protein = RelationshipSet(
'CODES', ['Transcript'], ['Protein'], ['sid'], ['sid'],
default_props={'source': 'ensembl'})
def run_with_mounted_arguments(self):
self.run(self.taxid)
def run(self, taxid):
ensembl_instance = self.get_instance_by_name('Ensembl')
datasource_name = ensembl_instance.datasource.name
# try patched path, if not available take flat
ensembl_gtf_file_path = Ensembl.get_gtf_file_path(taxid,
ensembl_instance,
patched=True)
if not os.path.exists(ensembl_gtf_file_path):
ensembl_gtf_file_path = Ensembl.get_gtf_file_path(taxid,
ensembl_instance,
patched=False)
annotation = GffReader(ensembl_gtf_file_path)
check_gene_ids = set()
check_transcript_ids = set()
check_protein_ids = set()
check_gene_transcript_rels = set()
check_transcript_protein_rels = set()
log.info("Start parsing ENSEMBL gtf file, taxid {}, {}".format(
taxid, ensembl_gtf_file_path))
for r in annotation.records:
# add gene node
gene_id = r.attributes['gene_id']
if gene_id not in check_gene_ids:
props = {
'sid': gene_id,
'name': r.attributes['gene_name'],
'taxid': taxid
}
self.genes.add_node(props)
check_gene_ids.add(gene_id)
# add transcript node
if r.type == 'transcript':
transcript_id = r.attributes['transcript_id']
if transcript_id not in check_transcript_ids:
props = {'sid': transcript_id, 'taxid': taxid}
self.transcripts.add_node(props)
check_transcript_ids.add(transcript_id)
# add protein node
if r.type == 'CDS':
protein_id = r.attributes['protein_id']
if protein_id not in check_protein_ids:
props = {'sid': protein_id, 'taxid': taxid}
self.proteins.add_node(props)
check_protein_ids.add(protein_id)
# Gene-CODES-Transcript
if r.type == 'transcript':
transcript_id = r.attributes['transcript_id']
gene_id = r.attributes['gene_id']
# add gene-transcript rel
if gene_id + transcript_id not in check_gene_transcript_rels:
self.gene_codes_transcript.add_relationship(
{'sid': gene_id}, {'sid': transcript_id}, {})
check_gene_transcript_rels.add(gene_id + transcript_id)
# Transcript-CODES-Protein
if r.type == 'CDS':
protein_id = r.attributes['protein_id']
transcript_id = r.attributes['transcript_id']
# add transcript-protein rel
if transcript_id + protein_id not in check_transcript_protein_rels:
self.transcript_codes_protein.add_relationship(
{'sid': transcript_id}, {'sid': protein_id}, {})
check_transcript_protein_rels.add(transcript_id +
protein_id)
log.info("Finished parsing ENSEMBL gtf file.")
