def test_Abundance(self):
"""Test canonicalization of abundances."""
short = Abundance(namespace='CHEBI', name='water')
self.assertEqual('a(CHEBI:water)', str(short))
long = Abundance(namespace='CHEBI', name='test name')
self.assertEqual('a(CHEBI:"test name")', str(long))
def test_str_has_both(self):
namespace, identifier = n(), n()
node = Abundance(namespace=namespace, identifier=identifier)
self.assertEqual(
'a({namespace}:{identifier})'.format(namespace=namespace, identifier=ensure_quotes(identifier)),
node.as_bel(),
)
def test_str_has_identifier(self):
namespace, identifier = n(), n()
node = Abundance(namespace=namespace, identifier=identifier)
self.assertEqual(
'a({namespace}:{identifier})'.format(namespace=namespace,
identifier=identifier),
node.as_bel())
def setUp(self):
self.graph: BELGraph = BELGraph()
test_chem: Abundance = Abundance('INCHIKEY', 'AAAAZQPHATYWOK-YRBRRWAQNA-N')
test_chem2: Abundance = Abundance('INCHIKEY', 'AAABHMIRDIOYOK-NPVYFSBINA-N')
self.graph.add_node_from_data(test_chem)
self.graph.add_node_from_data(test_chem2)
test_target: Protein = Protein('EGID', '2740')
test_target2: Protein = Protein('EGID', '2778')
self.graph.add_node_from_data(test_target)
self.graph.add_node_from_data(test_target2)
def test_simple(self):
"""Test converting a simple dictionary."""
namespace, name, identifier = n(), n(), n()
self.assertEqual(
Abundance(namespace=namespace, name=name),
_simple_po_to_dict({
FUNCTION: ABUNDANCE,
CONCEPT: {
NAMESPACE: namespace,
NAME: name,
},
}))
self.assertEqual(
Abundance(namespace=namespace, name=name, identifier=identifier),
_simple_po_to_dict({
FUNCTION: ABUNDANCE,
CONCEPT: {
NAMESPACE: namespace,
NAME: name,
IDENTIFIER: identifier,
},
}))
self.assertEqual(
Abundance(namespace=namespace, identifier=identifier),
_simple_po_to_dict({
FUNCTION: ABUNDANCE,
CONCEPT: {
NAMESPACE: namespace,
IDENTIFIER: identifier,
},
}))
with self.assertRaises(ValueError):
_simple_po_to_dict({
FUNCTION: ABUNDANCE,
CONCEPT: {
NAMESPACE: namespace,
},
})
def as_bel(self) -> Optional[BaseEntity]:
"""Convert this term to a BEL node."""
if self.namespace == 'biological_process':
return gobp(
name=self.name,
identifier=self.go_id,
)
if self.namespace == 'cellular_component':
if self.is_complex:
return NamedComplexAbundance(
namespace='go',
name=self.name,
identifier=self.go_id,
)
else:
return Abundance(
namespace='go',
name=self.name,
identifier=self.go_id,
)
def test_reaction(self):
"""Add identified reaction."""
graph = BELGraph()
reaction = Reaction(
namespace='rhea',
identifier='44104',
reactants=[
Abundance(namespace='chebi', identifier='17478'),
Abundance(namespace='chebi', identifier='15377'),
Abundance(namespace='chebi', identifier='57540'),
],
products=[
Abundance(namespace='chebi', identifier='29067'),
Abundance(namespace='chebi', identifier='15378'),
Abundance(namespace='chebi', identifier='57945'),
],
)
graph.add_node_from_data(reaction)
self.assertEqual(7, graph.number_of_nodes())
self.assertEqual(6, graph.number_of_edges())
adgrb_complex = ComplexAbundance([adgrb1, adgrb2])
achlorhydria = Pathology(namespace='MESHD', name='Achlorhydria')
akt1_rna = akt1.get_rna()
akt1_gene = akt1_rna.get_gene()
akt_methylated = akt1_gene.with_variants(GeneModification('Me'))
akt1_phe_508_del = akt1_gene.with_variants(Hgvs('p.Phe508del'))
cftr = hgnc('CFTR')
cftr_protein_unspecified_variant = cftr.with_variants(HgvsUnspecified())
cftr_protein_phe_508_del = cftr.with_variants(Hgvs('p.Phe508del'))
adenocarcinoma = Pathology('MESHD', 'Adenocarcinoma')
interleukin_23_complex = NamedComplexAbundance('GO', 'interleukin-23 complex')
oxygen_atom = Abundance(namespace='CHEBI', name='oxygen atom')
hydrogen_peroxide = Abundance('CHEBI', 'hydrogen peroxide')
tmprss2_gene = Gene('HGNC', 'TMPRSS2')
tmprss2_erg_gene_fusion = GeneFusion(
partner_5p=tmprss2_gene,
range_5p=EnumeratedFusionRange('c', 1, 79),
partner_3p=Gene('HGNC', 'ERG'),
range_3p=EnumeratedFusionRange('c', 312, 5034)
)
bcr_jak2_gene_fusion = GeneFusion(
partner_5p=Gene('HGNC', 'BCR'),
range_5p=EnumeratedFusionRange('c', '?', 1875),
partner_3p=Gene('HGNC', 'JAK2'),
from pybel.dsl import (
Abundance,
BiologicalProcess,
ComplexAbundance,
NamedComplexAbundance,
Pathology,
Protein,
ProteinModification,
)
from pybel.language import activity_mapping
from pybel.testing.constants import test_jgif_path
from tests.constants import TestGraphMixin
logging.getLogger('pybel.parser').setLevel(20)
calcium = Abundance('SCHEM', 'Calcium')
calcineurin_complex = NamedComplexAbundance('SCOMP', 'Calcineurin Complex')
foxo3 = Protein('HGNC', 'FOXO3')
tcell_proliferation = BiologicalProcess(
'GO', 'CD8-positive, alpha-beta T cell proliferation')
il15 = Protein('HGNC', 'IL15')
il2rg = Protein('MGI', 'Il2rg')
jgif_expected_nodes = {
calcium,
calcineurin_complex,
foxo3,
tcell_proliferation,
il15,
il2rg,
Protein('HGNC', 'CXCR6'),
Protein('HGNC', 'IL15RA'),
from pybel.typing import EdgeData
def _rel(x):
return {RELATION: x}
def _rela(x, y=None):
return {RELATION: x, OBJECT: activity(y)}
def _assoc(y):
return {RELATION: ASSOCIATION, 'association_type': y}
a1 = Abundance('CHEBI', '1')
p1 = Protein('HGNC', '1')
pf1 = Protein('INTERPRO', '1')
d1 = Pathology('MESH', '1')
b1 = BiologicalProcess('GO', '1')
b2 = BiologicalProcess('GO', '2')
m1 = MicroRna('MIRBASE', '1')
r1 = Rna('HGNC', '1')
r2 = Rna('HGNC', '2')
nca1 = NamedComplexAbundance('FPLX', '1')
pop1 = Population('taxonomy', '1')
p2 = Protein('HGNC', identifier='9236')
p3 = Protein('HGNC', identifier='9212')
r3 = p3.get_rna()
g3 = r3.get_gene()
def get_neurommsig_bel(
df: pd.DataFrame,
disease: str,
nift_values: Mapping[str, str],
) -> BELGraph:
"""Generate the NeuroMMSig BEL graph.
:param df:
:param disease:
:param nift_values: a dictionary of lower-cased to normal names in NIFT
"""
missing_features = set()
fixed_caps = set()
nift_value_originals = set(nift_values.values())
graph = BELGraph(
name=f'NeuroMMSigDB for {disease}',
description=f'SNP and Clinical Features for Subgraphs in {disease}',
authors=
'Daniel Domingo-Fernández, Charles Tapley Hoyt, Mufassra Naz, Aybuge Altay, Anandhi Iyappan',
contact='*****@*****.**',
version=time.strftime('%Y%m%d'),
)
for pathway, pathway_df in df.groupby(PATHWAY_COLUMN_NAME):
sorted_pathway_df = pathway_df.sort_values(GENE_COLUMN_NAME)
sliced_df = sorted_pathway_df[columns].itertuples()
for _, gene, pubmeds, lit_snps, gwas_snps, ld_block_snps, clinical_features, clinical_snps in sliced_df:
gene = ensure_quotes(gene)
for snp in itt.chain(lit_snps or [], gwas_snps or [], ld_block_snps
or [], clinical_snps or []):
if not snp.strip():
continue
graph.add_association(
Gene('HGNC', gene),
Gene('DBSNP', snp),
evidence=CANNED_EVIDENCE,
citation=CANNED_CITATION,
annotations={
'MeSHDisease': disease,
},
)
for clinical_feature in clinical_features or []:
if not clinical_feature.strip():
continue
if clinical_feature.lower() not in nift_values:
missing_features.add(clinical_feature)
continue
if clinical_feature not in nift_value_originals:
fixed_caps.add((clinical_feature,
nift_values[clinical_feature.lower()]))
clinical_feature = nift_values[
clinical_feature.lower()] # fix capitalization
graph.add_association(
Gene('HGNC', gene),
Abundance('NIFT', clinical_feature),
evidence=CANNED_EVIDENCE,
citation=CANNED_CITATION,
annotations={
'MeSHDisease': disease,
},
)
if clinical_snps:
for clinical_snp in clinical_snps:
graph.add_association(
Gene('DBSNP', clinical_snp),
Abundance('NIFT', clinical_feature),
evidence=CANNED_EVIDENCE,
citation=CANNED_CITATION,
annotations={
'MeSHDisease': disease,
},
)
if missing_features:
logger.warning('Missing Features in %s', disease)
for feature in missing_features:
logger.warning(feature)
if fixed_caps:
logger.warning('Fixed capitalization')
for broken, fixed in fixed_caps:
logger.warning('%s -> %s', broken, fixed)
return graph
def as_cas_bel(self) -> Abundance:
"""Get this drug as a PyBEL abundance identified by its CAS identifier."""
# https://www.ebi.ac.uk/miriam/main/datatypes/MIR:00000237
return Abundance(namespace='cas', identifier=self.cas_number)
# -*- coding: utf-8 -*-
from bio2bel_hmdb.enrich import *
from pybel import BELGraph
from pybel.dsl import Abundance, Pathology, Protein
from tests.constants import DatabaseMixin
hmdb_tuple1 = Abundance('HMDB', 'HMDB00008')
protein_tuple = Protein('UP', 'P50440')
# test enriching with tissues
hmdb_tuple2 = Abundance('HMDB', 'HMDB00064')
disease_tuple = Pathology('HMDB_D', 'Lung Cancer')
class TestEnrich(DatabaseMixin):
def test_enrich_metabolites_proteins(self):
g = BELGraph()
g.add_node_from_data(hmdb_tuple1)
self.assertEqual(1, g.number_of_nodes())
self.assertEqual(0, g.number_of_edges())
enrich_metabolites_proteins(g, self.manager)
self.assertEqual(4, g.number_of_nodes())
self.assertEqual(3, g.number_of_edges())
self.assertTrue(g.has_edge(protein_tuple, hmdb_tuple1))
def test_enrich_metabolites_diseases(self):
g = BELGraph()
g.add_node_from_data(hmdb_tuple2)
def test_abundance_as_bel(self):
"""Test converting an abundance to BEL with a name that does not need quotation."""
namespace, name = 'HGNC', 'YFG'
node = Abundance(namespace=namespace, name=name)
self.assertEqual('a(HGNC:YFG)', node.as_bel())
def as_drugbank_bel(self) -> Abundance:
"""Get this drug as a PyBEL abundance identified by its DrugBank identifier."""
return Abundance(namespace=MODULE_NAME,
name=self.name,
identifier=self.drugbank_id)
def test_abundance_as_bel_quoted(self):
"""Test converting an abundance to BEL with a name that needs quotation."""
namespace, name = 'HGNC', 'YFG-1'
node = Abundance(namespace=namespace, name=name)
self.assertEqual('a(HGNC:"YFG-1")', node.as_bel())
def test_as_tuple(self):
namespace, name = n(), n()
node = Abundance(namespace=namespace, name=name)
self.assertEqual(hash(node), hash(node.as_bel()))
def test_abundance_as_no_quotes(self):
"""Test converting an abundance that doesn't need quotes, but looks crazy."""
namespace, name = 'a-c', 'd.e.f'
node = Abundance(namespace=namespace, name=name)
self.assertEqual('a(a-c:d.e.f)', node.as_bel())
g1 = Gene(namespace=HGNC, name='1')
r1 = Rna(namespace=HGNC, name='1')
p1 = Protein(HGNC, name='1')
g2 = Gene(HGNC, name='2')
r2 = Rna(HGNC, name='2')
p2 = Protein(HGNC, name='2')
g3 = Gene(namespace=HGNC, name='3')
r3 = Rna(namespace=HGNC, name='3')
p3 = Protein(namespace=HGNC, name='3')
g4 = Gene(namespace=HGNC, name='4')
m4 = MicroRna(namespace=HGNC, name='4')
a5 = Abundance(namespace=CHEBI, name='5')
p5 = Pathology(namespace=GO, name='5')
class TestCollapseProteinInteractions(unittest.TestCase):
def test_protein_interaction_1(self):
graph = BELGraph()
graph.add_node_from_data(p1)
graph.add_node_from_data(p2)
graph.add_node_from_data(a5)
graph.add_node_from_data(p5)
graph.add_qualified_edge(p1, p2, relation=POSITIVE_CORRELATION, citation=n(), evidence=n())
graph.add_qualified_edge(p1, p2, relation=INCREASES, citation=n(), evidence=n())
def as_smiles_bel(self) -> Abundance:
"""Get this drug as a PyBEL abundance identified by SMILES."""
return Abundance(namespace='smiles', identifier=self.smiles)
def test_reaction(self):
node = Reaction(reactants=[Abundance(namespace='CHEBI', name='A')],
products=[Abundance(namespace='CHEBI', name='B')])
self.assertEqual('rxn(reactants(a(CHEBI:A)), products(a(CHEBI:B)))',
str(node))
def as_inchikey_bel(self) -> Abundance:
"""Get this drug as a PyBEL abundance identified by InChI-key."""
# https://www.ebi.ac.uk/miriam/main/datatypes/MIR:00000387
return Abundance(namespace='inchikey', identifier=self.inchikey)
def as_pubchem_compound_bel(self) -> Abundance:
"""Get this drug as a PyBEL abundance identified by PubChem."""
return Abundance(namespace='pubchem.compound',
identifier=self.pubchem_compound_id)
def normalize_graph_names(graph: BELGraph, database: str) -> None:
"""Normalize graph names."""
# Victim to Survivor (one to one node) mapping
one_to_one_mapping = {}
# Victim to Survivors (one to many nodes) mapping
one_to_many_mapping = defaultdict(set)
for node in graph.nodes():
# Skip ListAbundances and Reactions since they do not have a name
if isinstance(node, ListAbundance) or isinstance(
node, Reaction) or not node.name:
continue
# Normalize names: Lower case name and strip quotes or white spaces
lower_name = node.name.lower().strip('"').strip()
# Dealing with Genes/miRNAs
if isinstance(node, CentralDogma):
##################
# miRNA entities #
##################
if lower_name.startswith("mir"):
# Reactome preprocessing to flat multiple identifiers
if database == REACTOME:
reactome_cell = munge_reactome_gene(lower_name)
if isinstance(reactome_cell, list):
for lower_name in reactome_cell:
one_to_many_mapping[node].add(
MicroRna(
node.namespace,
name=lower_name.replace("mir-", "mir"),
identifier=node.identifier,
), )
if lower_name.endswith(' genes'):
lower_name = lower_name[:-len(' genes')]
elif lower_name.endswith(' gene'):
lower_name = lower_name[:-len(' gene')]
one_to_one_mapping[node] = MicroRna(
node.namespace,
name=lower_name.replace(
"mir-", "mir"), # Special case for Reactome
)
continue
# KEGG and Reactome
one_to_one_mapping[node] = MicroRna(
node.namespace,
name=node.name.replace("mir-", "mir"),
identifier=node.identifier,
)
##################
# Genes entities #
##################
else:
# Reactome preprocessing to flat multiple identifiers
if database == REACTOME:
reactome_cell = munge_reactome_gene(lower_name)
if isinstance(reactome_cell, list):
for lower_name in reactome_cell:
if lower_name in BLACK_LIST_REACTOME: # Filter entities in black list
continue
elif lower_name.startswith(
"("): # remove redundant parentheses
lower_name = lower_name.strip("(").strip(")")
one_to_many_mapping[node].add(
Protein(node.namespace,
name=lower_name,
identifier=node.identifier), )
else:
one_to_one_mapping[node] = Protein(
node.namespace,
name=lower_name,
identifier=node.identifier)
continue
# WikiPathways and KEGG do not require any processing of genes
elif database == WIKIPATHWAYS and lower_name in WIKIPATHWAYS_BIOL_PROCESS:
one_to_one_mapping[node] = BiologicalProcess(
node.namespace,
name=lower_name,
identifier=node.identifier,
)
continue
one_to_one_mapping[node] = Protein(node.namespace,
name=lower_name,
identifier=node.identifier)
#######################
# Metabolite entities #
#######################
elif isinstance(node, Abundance):
if database == 'wikipathways':
# Biological processes that are captured as abundance in
# BEL since they were characterized wrong in WikiPathways
if lower_name in WIKIPATHWAYS_BIOL_PROCESS:
one_to_one_mapping[node] = BiologicalProcess(
node.namespace,
name=lower_name,
identifier=node.identifier,
)
continue
# Abundances to BiologicalProcesses
elif (node.namespace
in {'WIKIDATA', 'WIKIPATHWAYS', 'REACTOME'}
and lower_name not in WIKIPATHWAYS_METAB):
one_to_one_mapping[node] = BiologicalProcess(
node.namespace,
name=lower_name,
identifier=node.identifier,
)
continue
# Fix naming in duplicate entity
if lower_name in WIKIPATHWAYS_NAME_NORMALIZATION:
lower_name = WIKIPATHWAYS_NAME_NORMALIZATION[lower_name]
elif database == REACTOME:
# Curated proteins that were coded as metabolites
if lower_name in REACTOME_PROT:
one_to_one_mapping[node] = Protein(
node.namespace,
name=lower_name,
identifier=node.identifier,
)
continue
# Flat multiple identifiers (this is not trivial because most of ChEBI names contain commas,
# so a clever way to fix some of the entities is to check that all identifiers contain letters)
elif "," in lower_name and all(
string.isalpha() for string in lower_name.split(",")):
for string in lower_name.split(","):
one_to_many_mapping[node].add(
Abundance(node.namespace,
name=string,
identifier=node.identifier), )
continue
one_to_one_mapping[node] = Abundance(node.namespace,
name=lower_name,
identifier=node.identifier)
#################################
# Biological Processes entities #
#################################
elif isinstance(node, BiologicalProcess):
# KEGG normalize name by removing the title prefix
if lower_name.startswith('title:'):
lower_name = lower_name[len('title:'):]
one_to_one_mapping[node] = BiologicalProcess(
node.namespace,
name=lower_name,
identifier=node.identifier,
)
relabel_nodes(graph, one_to_one_mapping)
multi_relabel(graph, one_to_many_mapping)