def __get_uniprot_reactions(self):
"""
Purpose: Takes reactions in the form of 'HSAXXX HSAXXX ReactionType HSAPathway and
converts each hsaXXX id to a swissprot uniprot id. Stores results inside a set of tuples
with the form (uprotA, uprotB, reaction label).
@return: Set of 4-tuples with the form (uprotA, uprotB, reaction label, direction).
"""
if self.__verbose:
print("Converting KEGG reactions to UniProt reactions...")
reaction_set = set()
for reaction in self.kegg_reactions:
hsa1 = reaction[0]
hsa2 = reaction[1]
reaction_type = reaction[2]
if hsa1 not in self.hsa_to_uprot.keys() \
or hsa2 not in self.hsa_to_uprot.keys():
continue
else:
u_prot_1 = self.hsa_to_uprot[hsa1]
u_prot_2 = self.hsa_to_uprot[hsa2]
ppi = PPI(u_prot_1, u_prot_2, [reaction_type])
if u_prot_1 != u_prot_2:
reaction_set.add(ppi)
if self.__verbose:
print("\tReaction: {}\t{}\t{}".format(
ppi.p1, ppi.p2, ppi.get_reaction_type_string()
))
reaction_set = set([(ppi.p1, ppi.p2, ppi.get_reaction_type_string()) for ppi in reaction_set])
return reaction_set
exclude = [
# None yet
]
ppis = []
n_ptms = []
unknown_ppis = []
unknown_n_ptms = []
for ptm in ptms:
clause_1 = (ptm.modification_type not in exclude) and ptm.has_enzyme()
clause_2 = ptm.enzyme_hprd_id != ptm.substrate_hprd_id
clause_3 = ptm.substrate_hprd_id != '-'
if clause_1 and clause_2 and clause_3:
ppi = PPI(
ptm.substrate_hprd_id,
ptm.enzyme_hprd_id,
ptm.modification_type.strip().lower().replace(' ', '-').split(',')
)
ppis.append(ppi)
n_ptms.append(ptm)
else:
ppi = PPI(
ptm.substrate_hprd_id,
ptm.enzyme_hprd_id,
ptm.modification_type.strip().lower().replace(' ', '-').split(',')
)
unknown_ppis.append(ppi)
unknown_n_ptms.append(ptm)
# Since there's a many swissprot to hprd_id mapping, priortise P, Q and O.
t = {'P':0, 'Q':1, 'O':2}