def load_data(self):
'''
- Loads all reactions, including locations for enzymes.
- Separates out the transport reactions as an class dictionary
- Makes mappings from molecule ids with location tags to external molecules without location tags
'''
# use rate_law_utilities to get all_reactions
all_reactions = load_reactions()
# make dict of reactions in TRANSPORT_IDS_FILE
self.all_transport_reactions = {}
rows = load_tsv(TRANSPORT_IDS_FILE)
for row in rows:
reaction_id = row['reaction id']
stoichiometry = all_reactions[reaction_id]['stoichiometry']
reversible = all_reactions[reaction_id]['is reversible']
transporters_loc = all_reactions[reaction_id]['catalyzed by']
self.all_transport_reactions[reaction_id] = {
'stoichiometry': stoichiometry,
'is reversible': reversible,
'catalyzed by': transporters_loc}
# Make map of external molecule_ids with a location tag (as used in reaction stoichiometry) to molecule_ids in the environment
self.molecule_to_external_map = {}
self.external_to_molecule_map = {}
rows = load_tsv(EXTERNAL_MOLECULES_FILE)
for row in rows:
molecule_id = row['molecule id']
location = row['exchange molecule location']
self.molecule_to_external_map[molecule_id + location] = molecule_id
self.external_to_molecule_map[molecule_id] = molecule_id + location
def load_reactions():
'''
Load all reactions, including the locations of enzymes into each reaction's 'catalyzed by' key
Returns:
all_reactions(dict)
'''
# get protein locations
proteins_locations = {}
rows = load_tsv(PROTEINS_FILE)
for row in rows:
molecule_id = row["id"]
location = row["location"]
molecule_loc = ['{}[{}]'.format(molecule_id, loc) for loc in location]
proteins_locations[molecule_id] = molecule_loc
# get complex locations
rows = load_tsv(COMPLEXATION_FILE)
for row in rows:
stoichiometry = row["stoichiometry"]
for stoich in stoichiometry:
molecule_id = stoich['molecule']
location = stoich["location"]
molecule_loc = [
'{}[{}]'.format(molecule_id, loc) for loc in location
]
proteins_locations[molecule_id] = molecule_loc
# make dict of all reactions
all_reactions = {}
rows = load_tsv(REACTIONS_FILE)
for row in rows:
reaction_id = row["reaction id"]
stoichiometry = row["stoichiometry"]
reversible = row["is reversible"]
enzymes = row["catalyzed by"]
# get location
enzymes_loc = []
for enzyme in enzymes:
if enzyme in proteins_locations.keys():
enzymes_loc.extend(proteins_locations[enzyme])
all_reactions[reaction_id] = {
"stoichiometry": stoichiometry,
"is reversible": reversible,
"catalyzed by": enzymes_loc
}
return all_reactions
def _load_tsv(self, dir_name, file_name):
path = self
steps = file_name.split(os.path.sep)
for subPath in steps[:-1]:
if not hasattr(path, subPath):
setattr(path, subPath, DataStore())
path = getattr(path, subPath)
attrName = steps[-1].split(".")[0]
setattr(path, attrName, [])
file_path = os.path.join(dir_name, file_name)
rows = load_tsv(file_path)
setattr(path, attrName, [row for row in rows])
def load_lookup(filename):
''' load a file and pass back dicts with lookup_avg and lookup_dist'''
lookup_avg = {}
lookup_dist = {}
rows = load_tsv(filename)
for row in rows:
key = row.get('id')
avg = row.get('average')
dist = row.get('distribution')
# check if empty distribution
if dist:
dist = [float(value) for value in dist]
else:
dist = [0]
lookup_avg[key] = avg
lookup_dist[key] = dist
return lookup_avg, lookup_dist