def test_combine_duplicates():
raf = Agent('RAF1')
mek = Agent('MEK1')
erk = Agent('ERK2')
p1 = Phosphorylation(raf, mek, evidence=Evidence(text='foo'))
p2 = Phosphorylation(raf, mek, evidence=Evidence(text='bar'))
p3 = Phosphorylation(raf, mek, evidence=Evidence(text='baz'))
p4 = Phosphorylation(raf, mek, evidence=Evidence(text='beep'))
p5 = Phosphorylation(mek, erk, evidence=Evidence(text='foo2'))
p6 = Dephosphorylation(mek, erk, evidence=Evidence(text='bar2'))
p7 = Dephosphorylation(mek, erk, evidence=Evidence(text='baz2'))
p8 = Dephosphorylation(mek, erk, evidence=Evidence(text='beep2'))
p9 = Dephosphorylation(Agent('SRC'),
Agent('KRAS'),
evidence=Evidence(text='beep'))
stmts = [p1, p2, p3, p4, p5, p6, p7, p8, p9]
pa = Preassembler(hierarchies, stmts=stmts)
pa.combine_duplicates()
# The statements come out sorted by their matches_key
assert len(pa.unique_stmts) == 4, len(pa.unique_stmts)
num_evs = [len(s.evidence) for s in pa.unique_stmts]
assert pa.unique_stmts[0].matches(p6) # MEK dephos ERK
assert num_evs[0] == 3, num_evs[0]
assert pa.unique_stmts[1].matches(p9) # SRC dephos KRAS
assert num_evs[1] == 1, num_evs[1]
assert pa.unique_stmts[2].matches(p5) # MEK phos ERK
assert num_evs[2] == 1, num_evs[2]
assert pa.unique_stmts[3].matches(p1) # RAF phos MEK
assert num_evs[3] == 4, num_evs[3]
def test_find_contradicts_refinement():
ras = Agent('RAS', db_refs={'FPLX': 'RAS'})
kras = Agent('KRAS', db_refs={'HGNC': '6407'})
hras = Agent('HRAS', db_refs={'HGNC': '5173'})
st1 = Phosphorylation(Agent('x'), ras)
st2 = Dephosphorylation(Agent('x'), kras)
st3 = Dephosphorylation(Agent('x'), hras)
pa = Preassembler(hierarchies, [st1, st2, st3])
contradicts = pa.find_contradicts()
assert len(contradicts) == 2
for s1, s2 in contradicts:
assert {s1.uuid, s2.uuid} in ({st1.uuid,
st2.uuid}, {st1.uuid, st3.uuid})
def get_observed_stmts(target_agent, observed_agent_forms, fold_change):
stmts = []
for obsf in observed_agent_forms:
obs_agent = Agent(obsf.name, db_refs=obsf.db_refs)
# If the agent has a modification then we make Modification
# statements to check
if obsf.mods:
for mod in obsf.mods:
# Create a Phosphorylation statement corresponding to
# this drug/Ab pair
if fold_change < 1:
stmt = Phosphorylation(target_agent, obs_agent,
mod.residue, mod.position)
else:
stmt = Dephosphorylation(target_agent, obs_agent,
mod.residue, mod.position)
# Otherwise the observed change is in protein amounts so we make
# RegulateAmount statements
else:
if fold_change < 1:
stmt = IncreaseAmount(target_agent, obs_agent)
else:
stmt = DecreaseAmount(target_agent, obs_agent)
stmts.append(stmt)
return stmts
def _get_gk_model_indra():
kras = Agent('KRAS', db_refs={'HGNC': '6407', 'UP': 'P01116'})
braf = Agent('BRAF', db_refs={'HGNC': '1097', 'UP': 'P15056'})
pp2a = Agent('PPP2CA')
st1 = Phosphorylation(kras, braf)
st2 = Dephosphorylation(pp2a, braf)
stmts = [st1, st2]
stmts_json = json.dumps(stmts_to_json(stmts))
return stmts_json
def process_phosphorylation_statements(self):
"""Looks for Phosphorylation events in the graph and extracts them into
INDRA statements.
In particular, looks for a Positive_regulation event node with a child
Phosphorylation event node.
If Positive_regulation has an outgoing Cause edge, that's the subject
If Phosphorylation has an outgoing Theme edge, that's the object
If Phosphorylation has an outgoing Site edge, that's the site
"""
G = self.G
statements = []
pwcs = self.find_event_parent_with_event_child('Positive_regulation',
'Phosphorylation')
for pair in pwcs:
(pos_reg, phos) = pair
cause = self.get_entity_text_for_relation(pos_reg, 'Cause')
theme = self.get_entity_text_for_relation(phos, 'Theme')
print('Cause:', cause, 'Theme:', theme)
# If the trigger word is dephosphorylate or similar, then we
# extract a dephosphorylation statement
trigger_word = self.get_entity_text_for_relation(
phos, 'Phosphorylation')
if 'dephos' in trigger_word:
deph = True
else:
deph = False
site = self.get_entity_text_for_relation(phos, 'Site')
theme_node = self.get_related_node(phos, 'Theme')
assert (theme_node is not None)
evidence = self.node_to_evidence(theme_node, is_direct=False)
if theme is not None:
if deph:
statements.append(
Dephosphorylation(s2a(cause),
s2a(theme),
site,
evidence=evidence))
else:
statements.append(
Phosphorylation(s2a(cause),
s2a(theme),
site,
evidence=evidence))
return statements
def test_multiprocessing():
braf = Agent('BRAF', db_refs={'HGNC': '1097'})
mek1 = Agent('MAP2K1', db_refs={'HGNC': '6840'})
mek = Agent('MEK', db_refs={'FPLX': 'MEK'})
# Statements
p0 = Phosphorylation(braf, mek)
p1 = Phosphorylation(braf, mek1)
p2 = Phosphorylation(braf, mek1, position='218')
p3 = Phosphorylation(braf, mek1, position='222')
p4 = Phosphorylation(braf, mek1, 'serine')
p5 = Phosphorylation(braf, mek1, 'serine', '218')
p6 = Phosphorylation(braf, mek1, 'serine', '222')
p7 = Dephosphorylation(braf, mek1)
stmts = [p0, p1, p2, p3, p4, p5, p6, p7]
pa = Preassembler(hierarchies, stmts=stmts)
# Size cutoff set to a low number so that one group will run remotely
# and one locally
toplevel = pa.combine_related(return_toplevel=True,
poolsize=1,
size_cutoff=2)
assert len(toplevel) == 3, 'Got %d toplevel statements.' % len(toplevel)
