def add_metacyc_pathway(self,metacyc_pathway_id,generic = False):
biocyc.set_organism('meta')
ptw = biocyc.get(metacyc_pathway_id)
network = self.scrap_metacyc_path(ptw,generic)
self.convert_compounds_into_modelseedid(network)
pathway = network.get_pathway()
if pathway:
target_reaction = pathway[-1]
last_reaction = biocyc.get(target_reaction)
products = last_reaction.compounds_right
targets = []
for product in products:
modelseedids = self.__compoundsIdConverter.convert_dbID_into_modelSeedId("MetaCyc",product.id)
if modelseedids is not None:
targets.extend(modelseedids)
self.__pathways_map[ptw.id] = network
self.target_pathways_map[ptw.id]=targets
def get_pathway_by_taxid_and_target(self,taxid,target):
paths_tax_range = []
taxa = NCBITaxa()
lineage = taxa.get_lineage(taxid)
for path in self.target_pathways_map:
pathway = biocyc.get(path)
taxonomic_range = pathway.taxonomic_range
for s in taxonomic_range:
taxid = int(s.id.replace("TAX-", ""))
if taxid in lineage:
paths_tax_range.append(path)
pathways = self.get_pathway_by_target(target)
for p in paths_tax_range:
if p in pathways:
return p
def setupSection(self, label, pathway_config):
# SHOW PATHWAYS
gb = QGroupBox(label)
vbox = QVBoxLayout()
# Populate the list boxes
self.label[label]['lw_pathways'] = ui.QBioCycPathwayTreeWidget([
'Biosynthesis',
'Degradation',
'Energy-Metabolism',
])
fwd_map = lambda x: biocyc.find_pathway_by_name(x).id
rev_map = lambda x: biocyc.get(x).name
self.config.hooks['QBioCycPathwayTreeWidget'] = self.config.hooks[
'QCheckTreeWidget'] # Works the same
self.config.add_handler(pathway_config,
self.label[label]['lw_pathways'],
(fwd_map, rev_map))
self.label[label]['lw_regExp'] = QLineEdit()
vbox.addWidget(self.label[label]['lw_pathways'])
vbox.addWidget(QLabel('Select/deselect matching pathways by name:'))
vboxh = QHBoxLayout()
vboxh.addWidget(self.label[label]['lw_regExp'])
addfr = QPushButton('-')
addfr.clicked.connect(self.onRegexpRemove)
addfr.setObjectName(label)
addfr.setFixedWidth(24)
remfr = QPushButton('+')
remfr.clicked.connect(self.onRegexpAdd)
remfr.setObjectName(label)
remfr.setFixedWidth(24)
vboxh.addWidget(addfr)
vboxh.addWidget(remfr)
vbox.addLayout(vboxh)
gb.setLayout(vbox)
self.layout.addWidget(gb)
def setupSection(self, label, pathway_config):
# SHOW PATHWAYS
gb = QGroupBox(label)
vbox = QVBoxLayout()
# Populate the list boxes
self.label[label]['lw_pathways'] = QListWidget()
self.label[label]['lw_pathways'].setSelectionMode(QAbstractItemView.ExtendedSelection)
self.label[label]['lw_pathways'].addItems(self.all_pathways)
self.label[label]['lw_pathways'].sortItems(Qt.AscendingOrder)
fwd_map = lambda x: biocyc.find_pathway_by_name(x).id
rev_map = lambda x: biocyc.get(x).name
self.config.add_handler(pathway_config, self.label[label]['lw_pathways'], (fwd_map, rev_map))
#for p in selected_pathways:
# self.label[label]['lw_pathways'].findItems(p, Qt.MatchExactly)[0].setSelected(True)
self.label[label]['lw_regExp'] = QLineEdit()
vbox.addWidget(self.label[label]['lw_pathways'])
vbox.addWidget(QLabel('Select/deselect matching pathways by name:'))
vboxh = QHBoxLayout()
vboxh.addWidget(self.label[label]['lw_regExp'])
addfr = QPushButton('-')
addfr.clicked.connect(self.onRegexpRemove)
addfr.setObjectName(label)
addfr.setFixedWidth(24)
remfr = QPushButton('+')
remfr.clicked.connect(self.onRegexpAdd)
remfr.setObjectName(label)
remfr.setFixedWidth(24)
vboxh.addWidget(addfr)
vboxh.addWidget(remfr)
vbox.addLayout(vboxh)
gb.setLayout(vbox)
self.layout.addWidget(gb)
def setupSection(self, label, pathway_config):
# SHOW PATHWAYS
gb = QGroupBox(label)
vbox = QVBoxLayout()
# Populate the list boxes
self.label[label]['lw_pathways'] = ui.QBioCycPathwayTreeWidget([
'Biosynthesis',
'Degradation',
'Energy-Metabolism',
])
fwd_map = lambda x: biocyc.find_pathway_by_name(x).id
rev_map = lambda x: biocyc.get(x).name
self.config.hooks['QBioCycPathwayTreeWidget'] = self.config.hooks['QCheckTreeWidget'] # Works the same
self.config.add_handler(pathway_config, self.label[label]['lw_pathways'], (fwd_map, rev_map))
self.label[label]['lw_regExp'] = QLineEdit()
vbox.addWidget(self.label[label]['lw_pathways'])
vbox.addWidget(QLabel('Select/deselect matching pathways by name:'))
vboxh = QHBoxLayout()
vboxh.addWidget(self.label[label]['lw_regExp'])
addfr = QPushButton('-')
addfr.clicked.connect(self.onRegexpRemove)
addfr.setObjectName(label)
addfr.setFixedWidth(24)
remfr = QPushButton('+')
remfr.clicked.connect(self.onRegexpAdd)
remfr.setObjectName(label)
remfr.setFixedWidth(24)
vboxh.addWidget(addfr)
vboxh.addWidget(remfr)
vbox.addLayout(vboxh)
gb.setLayout(vbox)
self.layout.addWidget(gb)
def get_pathway_by_taxID(self,taxID):
# paths_species = []
paths_tax_range = []
taxa = NCBITaxa()
lineage = taxa.get_lineage(taxID)
for path in self.target_pathways_map:
pathway = biocyc.get(path)
# species = pathway.species
# for s in species:
# if re.search("TAX-",s.id):
# id = int(s.id.replace("TAX-", ""))
# if id in lineage:
# paths_species.append((path,pathway.name))
taxonomic_range = pathway.taxonomic_range
for s in taxonomic_range:
taxid = int(s.id.replace("TAX-",""))
if taxid in lineage:
paths_tax_range.append(path)
# res
return paths_tax_range
if input_data is not None:
datam = input_data.mean()
# We need BioCyc identifiers
if 'BioCyc' in input_data.columns.names:
if type(input_data.columns) == pd.MultiIndex:
entities = [k[input_data.columns.names.index('BioCyc')] for k in input_1.columns.values]
else:
entities = input_data.columns.values
# Map to BioCyc if not already
biocyc_entities = []
for e in entities:
if hasattr(e, 'id'):
biocyc_entities.append(e)
elif type(e) is str:
try:
biocyc_entities[n] = biocyc.get(o)
except:
biocyc_entities[n] = None
else:
biocyc_entities.append(None)
datas = [(e,s) for e,s in zip(biocyc_entities, datam.values) if e is not None]
data.extend(datas)
print "%d entities with data" % len(data)
results = mining(data, target=config['/Data/MiningTarget'],
include=config['include_pathways'],
exclude=config['exclude_pathways'],
no_of_results=config['/Data/MiningDepth'],
algorithm=config['/Data/MiningType'],
if 'BioCyc' in input_data.columns.names:
if type(input_data.columns) == pd.MultiIndex:
entities = [
k[input_data.columns.names.index('BioCyc')]
for k in input_1.columns.values
]
else:
entities = input_data.columns.values
# Map to BioCyc if not already
biocyc_entities = []
for e in entities:
if hasattr(e, 'id'):
biocyc_entities.append(e)
elif type(e) is str:
try:
biocyc_entities[n] = biocyc.get(o)
except:
biocyc_entities[n] = None
else:
biocyc_entities.append(None)
datas = [(e, s) for e, s in zip(biocyc_entities, datam.values)
if e is not None]
data.extend(datas)
print "%d entities with data" % len(data)
results = mining(
data,
target=config['/Data/MiningTarget'],
include=config['include_pathways'],
if suggested_pathways is not None:
# Pathways should come in as a column set named 'BioCyc' but containing Pathway objects
if 'BioCyc' in suggested_pathways.columns.names:
if type(suggested_pathways.columns) == pd.Index:
ps = suggested_pathways.columns.values
elif type(suggested_pathways.columns) == pd.MultiIndex:
ps = suggested_pathways.columns.values[
suggested_pathways.columns.names.index('BioCyc')]
pathways = []
for p in ps:
if type(p) == Pathway:
pathways.append(p)
else:
try:
p = biocyc.get(id)
except:
pass
else:
if type(p) == Pathway:
pathways.append(p)
else:
pathways = []
pathways.extend([biocyc.get(p) for p in config['show_pathways']])
pathways = [p for p in pathways if p.id not in config['hide_pathways']]
pathways = list(set(pathways))
print("Forcing display of %s" % config['show_pathways'])
print("Suppressing display of %s" % config['hide_pathways'])
count = 0
for n, l in enumerate(labels):
if current_biocyc[n] is not None:
continue
o = lku(l)
# Note that the BioCyc module returns an object, so we need to convert to ID if it is
if hasattr(o, 'id'):
current_biocyc[n] = o
count += 1
elif type(o) is str:
try:
current_biocyc[n] = biocyc.get(o)
except:
current_biocyc[n] = None
else:
count += 1
else:
current_biocyc[n] = None
print("Matched %d identifiers" % count)
cross_maps = {}
if 1 == 0:
for m, f in {'BiGG': 'bigg', 'HMDB': 'hmdb', 'KEGG': 'kegg'}.items():
lku = reverse_map_generator(f)
if suggested_pathways is not None:
# Pathways should come in as a column set named 'BioCyc' but containing Pathway objects
if 'BioCyc' in suggested_pathways.columns.names:
if type(suggested_pathways.columns) == pd.Index:
ps = suggested_pathways.columns.values
elif type(suggested_pathways.columns) == pd.MultiIndex:
ps = suggested_pathways.columns.values[ suggested_pathways.columns.names.index('BioCyc') ]
pathways = []
for p in ps:
if type(p) == Pathway:
pathways.append(p)
else:
try:
p = biocyc.get(id)
except:
pass
else:
if type(p) == Pathway:
pathways.append(p)
else:
pathways = []
pathways.extend( [biocyc.get(p) for p in config['show_pathways'] ])
pathways = [p for p in pathways if p.id not in config['hide_pathways']]
pathways = list( set( pathways ) )
print("Forcing display of %s" % config['show_pathways'])
print("Suppressing display of %s" % config['hide_pathways'])
fetch_queue = pathways = [
'Biosynthesis',
'Degradation',
'Energy-Metabolism',
]
fetched = []
added_this_loop = None
while len(fetch_queue) > 0:
added_this_loop = 0
to_add = []
for p in fetch_queue:
print p
pw = biocyc.get(p)
to_add.extend(pw._super_pathways)
to_add.extend(pw._subclasses)
to_add.extend(pw._instances)
to_add = [p for p in to_add if p not in fetched]
fetch_queue = to_add
fetched.extend(to_add)
DEFAULT_PATHWAYS = fetched
total_n = len(DEFAULT_PATHWAYS)
for n, p in enumerate(DEFAULT_PATHWAYS):
pw = biocyc.get(p)
print "P", pw, pw.id, "(%d/%d)" % (n, total_n)
for r in pw.reactions:
count = 0
for n, l in enumerate(labels):
if current_biocyc[n] is not None:
continue
o = lku(l)
# Note that the BioCyc module returns an object, so we need to convert to ID if it is
if hasattr(o, 'id'):
current_biocyc[n] = o
count += 1
elif type(o) is str:
try:
current_biocyc[n] = biocyc.get(o)
except:
current_biocyc[n] = None
else:
count += 1
else:
current_biocyc[n] = None
print("Matched %d identifiers" % count)
cross_maps = {}
if 1 == 0:
for m, f in {'BiGG': 'bigg', 'HMDB': 'hmdb', 'KEGG': 'kegg'}.items():
lku = reverse_map_generator(f)
cm = []
fetch_queue = pathways = [
'Biosynthesis',
'Degradation',
'Energy-Metabolism',
]
fetched = []
added_this_loop = None
while len(fetch_queue) > 0:
added_this_loop = 0
to_add = []
for p in fetch_queue:
print p
pw = biocyc.get(p)
to_add.extend(pw._super_pathways)
to_add.extend(pw._subclasses)
to_add.extend(pw._instances)
to_add = [p for p in to_add if p not in fetched]
fetch_queue = to_add
fetched.extend(to_add)
DEFAULT_PATHWAYS = fetched
total_n = len(DEFAULT_PATHWAYS)
for n, p in enumerate(DEFAULT_PATHWAYS):
pw = biocyc.get(p)
print "P", pw, pw.id, "(%d/%d)" % (n, total_n)
for r in pw.reactions: