def test_list_abundance_has_contents(self):
"""Test that the construction of list abundance doesn't have empty lists."""
with self.assertRaises(ListAbundanceEmptyException):
ComplexAbundance([])
with self.assertRaises(ListAbundanceEmptyException):
CompositeAbundance([])
def test_complex_with_name(self):
"""Test what happens with a named complex.
.. code-block::
complex(SCOMP:"9-1-1 Complex") hasComponent p(HGNC:HUS1)
complex(SCOMP:"9-1-1 Complex") hasComponent p(HGNC:RAD1)
complex(SCOMP:"9-1-1 Complex") hasComponent p(HGNC:RAD9A)
"""
hus1 = Protein(namespace='HGNC', name='HUS1')
rad1 = Protein(namespace='HGNC', name='RAD1')
rad9a = Protein(namespace='HGNC', name='RAD9A')
members = [hus1, rad1, rad9a]
nine_one_one = ComplexAbundance(members=members,
namespace='SCOMP',
name='9-1-1 Complex')
graph = BELGraph()
graph.add_node_from_data(nine_one_one)
self.assertIn(nine_one_one, graph)
self.assertIn(hus1, graph)
self.assertIn(rad1, graph)
self.assertIn(rad9a, graph)
def _tf_down(graph, protein, rna):
graph.add_directly_decreases(
ComplexAbundance([protein, rna.get_gene()]),
rna,
citation=n(),
evidence=n(),
)
def setUp(self) -> None:
"""Set up a small test graph."""
self.graph = BELGraph()
self.graph.add_increases(ComplexAbundance([p1, g2]),
r2,
citation=n(),
evidence=n())
self.graph.add_increases(p2, p3, citation=n(), evidence=n())
def test_complex(self):
node = ComplexAbundance(members=[fos, jun])
self.graph.add_node_from_data(node)
self.assertIn(node, self.graph)
self.assertEqual(3, self.graph.number_of_nodes())
self.assertEqual(2, self.graph.number_of_edges())
assert_unqualified_edge(self, fos, node, PART_OF)
assert_unqualified_edge(self, jun, node, PART_OF)
def test_binds(self):
"""Test the ``binds`` relation."""
statement = 'p(HGNC:X) binds p(HGNC:Y)'
self.parser.relation.parseString(statement)
source = Protein('HGNC', 'X')
target = Protein('HGNC', 'Y')
x_y_complex = ComplexAbundance([source, target])
self.assert_has_node(x_y_complex)
self.assert_has_edge(source, x_y_complex, relation=PART_OF)
self.assert_has_edge(target, x_y_complex, relation=PART_OF)
def test_named_complex(self):
x = ComplexAbundance(namespace='a',
identifier='b',
members=[
Protein(namespace='c', identifier='d'),
Protein(namespace='c', identifier='e'),
])
y = parse_result_to_dsl(dict(x))
self.assertIsInstance(y, ComplexAbundance)
self.assertIn(pc.MEMBERS, y)
self.assertIn(pc.CONCEPT, y)
def test_get_tf_pairs(self):
"""Test iterating over transcription factor pairs."""
graph = BELGraph()
p1, p2, p3 = (Protein('test', str(i)) for i in range(1, 4))
r4, r5, r6 = (Rna('test', str(j)) for j in range(4, 7))
g4 = r4.get_gene()
self.assertIsNotNone(g4)
g5 = r5.get_gene()
self.assertIsNotNone(g5)
c14, c25 = ComplexAbundance([p1, g4]), ComplexAbundance([p2, g5])
_tf_up(graph, p1, r4)
_tf_down(graph, p2, r5)
graph.add_correlation(p3, r6, citation=n(), evidence=n())
self.assertEqual({p1, p2, p3, r4, r5, r6, g4, g5, c14, c25},
set(graph))
expected_edges = [
(c14, r4),
(p1, c14),
(g4, c14),
(c25, r5),
(p2, c25),
(g5, c25),
(p3, r6),
(r6, p3),
]
sorted_expected_edges = sorted(expected_edges, key=_bel_pair_key)
sorted_actual_edges = sorted(graph.edges(), key=_bel_pair_key)
self.assertEqual(sorted_expected_edges, sorted_actual_edges)
pairs = set(get_tf_pairs(graph))
expected_pairs = {(p1, r4, +1), (p2, r5, -1)}
self.assertEqual(expected_pairs, pairs)
def bel_isolated_reconstituted(self, graph: BELGraph):
"""Run the isolated node test."""
self.assertIsNotNone(graph)
self.assertIsInstance(graph, BELGraph)
adgrb1 = Protein(namespace='HGNC', name='ADGRB1')
adgrb2 = Protein(namespace='HGNC', name='ADGRB2')
adgrb_complex = ComplexAbundance([adgrb1, adgrb2])
achlorhydria = Pathology(namespace='MESHD', name='Achlorhydria')
for node in graph:
self.assertIsInstance(node, BaseEntity)
self.assertIn(adgrb1, graph)
self.assertIn(adgrb2, graph)
self.assertIn(adgrb_complex, graph)
self.assertIn(achlorhydria, graph)
assert_has_edge(self, adgrb1, adgrb_complex, graph, relation=PART_OF)
assert_has_edge(self, adgrb2, adgrb_complex, graph, relation=PART_OF)
def test_import(self):
"""Test importing a hipathia network as a BEL graph."""
graph = from_hipathia_paths(
name='test1',
att_path=TEST_1_ATT_PATH,
sif_path=TEST_1_SIF_PATH,
)
a = Protein(namespace='ncbigene', identifier='1')
b_family = Protein(namespace='hipathia.family', identifier='B_Family')
b_2 = Protein(namespace='ncbigene', identifier='2')
b_3 = Protein(namespace='ncbigene', identifier='3')
c_family = Protein(namespace='hipathia.family', identifier='C_Family')
c_4 = Protein(namespace='ncbigene', identifier='4')
c_5 = Protein(namespace='ncbigene', identifier='5')
d = Protein(namespace='ncbigene', identifier='6')
c_d = ComplexAbundance([c_family, d])
self.assertEqual(
sorted({a, b_family, b_2, b_3, c_family, c_4, c_5, d, c_d},
key=str),
sorted(graph, key=str),
)
BelReconstitutionMixin,
akt1,
casp8,
egfr,
expected_test_simple_metadata,
fadd,
test_citation_dict,
test_evidence_text,
)
fos = hgnc(name='FOS')
jun = hgnc(name='JUN')
mirna_1 = mirbase(name=n())
mirna_2 = mirbase(name=n())
pathology_1 = Pathology('DO', n())
ap1_complex = ComplexAbundance([fos, jun])
egfr_dimer = ComplexAbundance([egfr, egfr])
yfg_data = hgnc(name='YFG')
e2f4_data = hgnc(name='E2F4')
bound_ap1_e2f4 = ComplexAbundance([ap1_complex, e2f4_data])
superoxide = chebi(name='superoxide')
hydrogen_peroxide = chebi(name='hydrogen peroxide')
oxygen = chebi(name='oxygen')
superoxide_decomposition = Reaction(reactants=[superoxide],
products=[hydrogen_peroxide, oxygen])
def assert_unqualified_edge(test_case, u: BaseEntity, v: BaseEntity,
def test_complex_abundance(self):
node = ComplexAbundance(members=[
Protein(namespace='HGNC', name='FOS'),
Protein(namespace='HGNC', name='JUN')
])
self.assertEqual('complex(p(HGNC:FOS), p(HGNC:JUN))', str(node))
pmod,
protein_fusion,
protein_substitution,
reaction,
rna,
)
from pybel.utils import citation_dict
example_graph = BELGraph()
ptk2 = Protein(namespace='hgnc', name='PTK2', variants=pmod('Ph', 'Tyr', 925))
mapk1 = Protein(namespace='hgnc', name='MAPK1')
mapk3 = Protein(namespace='hgnc', name='MAPK3')
grb2 = Protein(namespace='hgnc', name='GRB2')
sos1 = Protein(namespace='hgnc', name='SOS1')
ptk2_rgb2_sos1 = ComplexAbundance([mapk1, grb2, sos1])
ras_family = Protein(namespace='fplx', name='RAS')
pi3k_complex = named_complex_abundance(namespace='fplx',
name='p85/p110 PI3Kinase Complex')
kinase_activity = activity('kin')
catalytic_activity = activity('cat')
gtp_activity = activity('gtp')
c1 = '10446041'
e1 = "FAK also combines with, and may activate, phosphoinositide 3-OH kinase (PI 3-kinase), either directly or " \
"through the Src kinase (13). Finally, there is evidence that Src phosphorylates FAK at Tyr925, creating a" \
" binding site for the complex of the adapter Grb2 and Ras guanosine 5'-triphosphate exchange factor mSOS (10)." \
" These interactions link FAK to signaling pathways that modify the cytoskeleton and activate mitogen-activated" \
" protein kinase (MAPK) cascades (Fig. 3A)."
citation_2 = citation_dict(db=CITATION_TYPE_PUBMED, db_id='123456')
evidence_1 = "Evidence 1"
dummy_evidence = 'These are mostly made up'
akt1 = hgnc(name='AKT1')
egfr = hgnc(name='EGFR')
fadd = hgnc(name='FADD')
casp8 = hgnc(name='CASP8')
mia = hgnc(name='MIA')
il6 = Protein('HGNC', 'IL6')
adgrb1 = Protein(namespace='HGNC', name='ADGRB1')
adgrb2 = Protein(namespace='HGNC', name='ADGRB2')
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')
Protein('HGNC', 'LCK'),
Protein('SFAM', 'Chemokine Receptor Family'),
Protein('HGNC', 'CXCL9'),
Pathology('SDIS', 'T-cell migration'),
Protein('HGNC', 'CXCR3'),
Abundance('CHEBI', 'acrolein'),
Protein('HGNC', 'IDO2'),
Pathology('MESHD', 'Pulmonary Disease, Chronic Obstructive'),
Protein('HGNC', 'IFNG'),
Protein('HGNC', 'TNFRSF4'),
Protein('HGNC', 'CTLA4'),
Protein('HGNC', 'GZMA'),
Protein('HGNC', 'PRF1'),
Protein('HGNC', 'TNF'),
Protein('SFAM', 'Chemokine Receptor Family'),
ComplexAbundance([Protein('HGNC', 'CD8A'),
Protein('HGNC', 'CD8B')]),
ComplexAbundance([Protein('HGNC', 'CD8A'),
Protein('HGNC', 'CD8B')]),
Protein('HGNC', 'PLCG1', variants=ProteinModification('Ph', 'Tyr')),
Protein('EGID', '21577'),
}
jgif_expected_edges = [
(calcium, calcineurin_complex, {
RELATION: DIRECTLY_INCREASES,
EVIDENCE:
'NMDA-mediated influx of calcium led to activated of the calcium-dependent phosphatase calcineurin and the subsequent dephosphorylation and activation of the protein-tyrosine phosphatase STEP',
CITATION: {
NAMESPACE: CITATION_TYPE_PUBMED,
IDENTIFIER: '12483215'
},
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()
c1 = ComplexAbundance([p2, g3])
c2 = ComplexAbundance([p1, p2])
c3 = ComplexAbundance([a1, p2])
converters_true_list = [
(PartOfNamedComplexConverter, p1, nca1, _rel(PART_OF), ('HGNC:1', 'partOf',
'FPLX:1')),
(SubprocessPartOfBiologicalProcessConverter, b1, b2, _rel(PART_OF),
('GO:1', 'partOf', 'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(INCREASES), ('GO:1', INCREASES,
'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(DECREASES), ('GO:1', DECREASES,
'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(DIRECTLY_INCREASES),
('GO:1', DIRECTLY_INCREASES, 'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(DIRECTLY_DECREASES),
# -*- coding: utf-8 -*-
import unittest
from pybel import BELGraph
from pybel.dsl import ComplexAbundance, Fragment, Protein
from pybel.dsl.namespaces import hgnc
from pybel_tools.selection.group_nodes import get_mapped_nodes
ccl2 = hgnc(name='CCL2')
ccr2 = hgnc(name='CCR2')
ccl2_mgi = Protein('MGI', 'Ccl2')
ccl2_ccr2_complex = ComplexAbundance([ccl2, ccr2])
chemokine_family = Protein('FPLX', 'chemokine protein family')
HGNC = 'hgnc'
class TestMapping(unittest.TestCase):
def test_variants_mapping(self):
graph = BELGraph()
app = Protein(HGNC, 'APP')
app_fragment = app.with_variants(Fragment('1_49'))
graph.add_node_from_data(app_fragment)
mapped_nodes = get_mapped_nodes(graph, HGNC, {'APP'})
self.assertEqual(1, len(mapped_nodes))
self.assertIn(app, mapped_nodes)
self.assertEqual({app_fragment}, mapped_nodes[app])
PROTEIN_NAMESPACE = 'ncbigene'
FAMILY_NAMESPACE = 'hipathia.family'
a = Protein(namespace='ncbigene', identifier='P001', name='A')
b_family = Protein(namespace='hipathia.family',
identifier='F001',
name='B_Family')
b1 = Protein(namespace='ncbigene', identifier='P002', name='B1')
b2 = Protein(namespace='ncbigene', identifier='P003', name='B2')
c_family = Protein(namespace='hipathia.family',
identifier='F002',
name='C_Family')
c1 = Protein(namespace='ncbigene', identifier='P004', name='C1')
c2 = Protein(namespace='ncbigene', identifier='P005', name='C2')
d = Protein(namespace='ncbigene', identifier='P006', name='D')
c_d = ComplexAbundance([c_family, d])
e = Protein(namespace='ncbigene', identifier='P007', name='E')
f = Protein(namespace='ncbigene', identifier='P008', name='F')
e_f = ComplexAbundance([e, f])
name = 'test'
class TestExportHipathia(unittest.TestCase):
"""Test Hipathia."""
def setUp(self) -> None:
self.graph = BELGraph(name=name)
def _get_dfs(self) -> Tuple[pd.DataFrame, pd.DataFrame]:
return to_hipathia_dfs(self.graph)
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()
c1 = ComplexAbundance([p2, g3])
c2 = ComplexAbundance([p1, p2])
c3 = ComplexAbundance([a1, p2])
p1_homodimer = ComplexAbundance([p1, p1])
p1_homotrimer = ComplexAbundance([p1, p1, p1])
converters_true_list = [
(PartOfNamedComplexConverter, p1, nca1, _rel(PART_OF), ('HGNC:1', 'partOf',
'FPLX:1')),
(SubprocessPartOfBiologicalProcessConverter, b1, b2, _rel(PART_OF),
('GO:1', 'partOf', 'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(INCREASES), ('GO:1', INCREASES,
'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(DECREASES), ('GO:1', DECREASES,
'GO:2')),
(tsvc.ProcessCausalConverter, b1, b2, _rel(DIRECTLY_INCREASES),
def test_empty_complex(self):
"""Test that an empty complex causes a failure."""
with self.assertRaises(ValueError):
ComplexAbundance(members=[])