def init(self):
if self.db.DoesTableExist(self.THERMODYNAMICS_TABLE_NAME):
logging.info('Reading thermodynamic data from database')
reader = self.db.DictReader(self.THERMODYNAMICS_TABLE_NAME)
PsuedoisomerTableThermodynamics._FromDictReader(
reader, self, label=None, name="Unified Group Contribution",
warn_for_conflicting_refs=False)
conservation_rows = []
for row in self.db.DictReader(self.CONSERVATIONS_TABLE_NAME):
sparse = dict((int(cid), coeff) for (cid, coeff) in json.loads(row['json']).iteritems())
msg = row['msg']
conservation_rows.append((msg, sparse))
logging.info('Reading conservation matrix data from database')
all_cids = sorted(self.kegg.get_all_cids())
cid_dict = dict((cid, i) for (i, cid) in enumerate(all_cids))
self.P_L_tot = np.matrix(np.zeros((len(conservation_rows), len(all_cids))))
for i, (msg, sparse) in enumerate(conservation_rows):
for cid, coeff in sparse.iteritems():
if cid not in cid_dict:
raise Exception("ERROR: C%05d is not found in KEGG but appears in our database"
% cid)
self.P_L_tot[i, cid_dict[cid]] = float(coeff)
else:
self.LoadGroups(True)
self.LoadObservations(True)
self.LoadGroupVectors(True)
self.LoadData(True)
self.EstimateKeggCids()
def setUp(self):
fake_csv_file = StringIO(CSV_DATA)
csv_reader = csv.DictReader(fake_csv_file)
self.fake_thermo_csv = PsuedoisomerTableThermodynamics()
self.fake_thermo_csv = PsuedoisomerTableThermodynamics._FromDictReader(
csv_reader, self.fake_thermo_csv, warn_for_conflicting_refs=False)
db = SqliteDatabase(PUBLIC_DB_FNAME)
db_reader = db.DictReader('fake_pseudoisomers')
self.fake_thermo_db = PsuedoisomerTableThermodynamics()
self.fake_thermo_db = PsuedoisomerTableThermodynamics._FromDictReader(
db_reader, self.fake_thermo_db, warn_for_conflicting_refs=False)
def init(self):
self.LoadGroups(True)
self.LoadObservations(True)
self.LoadGroupVectors(True)
if self.db.DoesTableExist(self.CONTRIBUTION_TABLE_NAME):
self.LoadContributionsFromDB()
else:
self.Train()
self.EstimateKeggCids()
reader = self.db.DictReader(self.THERMODYNAMICS_TABLE_NAME)
PsuedoisomerTableThermodynamics._FromDictReader(
reader, self, label=None, name="Group Contribution",
warn_for_conflicting_refs=False)
def init(self):
self.LoadGroups(True)
self.LoadObservations(True)
self.LoadGroupVectors(True)
if self.db.DoesTableExist(self.CONTRIBUTION_TABLE_NAME):
self.LoadContributionsFromDB()
else:
self.Train()
self.EstimateKeggCids()
reader = self.db.DictReader(self.THERMODYNAMICS_TABLE_NAME)
PsuedoisomerTableThermodynamics._FromDictReader(
reader,
self,
label=None,
name="Group Contribution",
warn_for_conflicting_refs=False)
def init(self):
if self.db.DoesTableExist(self.THERMODYNAMICS_TABLE_NAME):
logging.info('Reading thermodynamic data from database')
reader = self.db.DictReader(self.THERMODYNAMICS_TABLE_NAME)
PsuedoisomerTableThermodynamics._FromDictReader(
reader,
self,
label=None,
name="Unified Group Contribution",
warn_for_conflicting_refs=False)
conservation_rows = []
for row in self.db.DictReader(self.CONSERVATIONS_TABLE_NAME):
sparse = dict(
(int(cid), coeff)
for (cid, coeff) in json.loads(row['json']).iteritems())
msg = row['msg']
conservation_rows.append((msg, sparse))
logging.info('Reading conservation matrix data from database')
all_cids = sorted(self.kegg.get_all_cids())
cid_dict = dict((cid, i) for (i, cid) in enumerate(all_cids))
self.P_L_tot = np.matrix(
np.zeros((len(conservation_rows), len(all_cids))))
for i, (msg, sparse) in enumerate(conservation_rows):
for cid, coeff in sparse.iteritems():
if cid not in cid_dict:
raise Exception(
"ERROR: C%05d is not found in KEGG but appears in our database"
% cid)
self.P_L_tot[i, cid_dict[cid]] = float(coeff)
else:
self.LoadGroups(True)
self.LoadObservations(True)
self.LoadGroupVectors(True)
self.LoadData(True)
self.EstimateKeggCids()
