def routine_2(fn, k, tree, numOfLines):
kmers_examined = 0
with open(fn, 'r') as fh:
count = 0
for myline in fh:
count += 1
print('seq = {0}, k = {1}'.format(count,k))
for j in range(len(myline) - k):
kmers_examined += 1
this_kmer = myline[j:j+k]
if count <= int(numOfLines * training_perc):
tree.find_in_tree(this_kmer, False , kmers_examined, k, True, sequenceIndex=count-1)
else:
tree.find_in_tree(this_kmer, True, kmers_examined, k, True, sequenceIndex=count-1)
TreeClass.found_kmers = False
TreeClass.check_tree(root, kmers_examined, k)
return tree
def add_all_nodes(current, depth):
chars = ['A', 'C', 'G', 'T']
for char in chars:
current.add_child(TreeClass.Node(char, current, depth+1))
return current
def predict(self, method, age, sex, days, criteri,rhytmc, rhytm):
if method == "Логистическая регрессия":
return LogReg.Predict(age, sex, days, criteri,rhytmc, rhytm)
if method == "Метод опорных векторов":
return SVMFIle.Predict(age, sex, days, criteri,rhytmc, rhytm)
if method == "Дерево решений":
return TreeClass.Predict(age, sex, days, criteri,rhytmc, rhytm)
if method == "Случайный лес":
return RandomForestFile.Predict(age, sex, days, criteri,rhytmc, rhytm)
def initialize_tree():
tree = TreeClass.Tree()
root = tree.root
root = add_all_nodes(root , 0)
for child1 in root.children:
child1 = add_all_nodes(child1, 1)
for child2 in child1.children:
child2 = add_all_nodes(child2, 2)
for child3 in child2.children:
child3 = add_all_nodes(child3, 3)
return root, tree
if __name__ == "__main__":
def log_uncaught_exceptions(ex_cls, ex, tb):
text = '{}: {}:\n'.format(ex_cls.__name__, ex)
import traceback
text += ''.join(traceback.format_tb(tb))
print(text)
QtWidgets.QMessageBox.critical(None, 'Error', text)
quit()
sys.excepthook = log_uncaught_exceptions
if not os.path.isfile('logReg') or not os.path.isfile('SVM') or not os.path.isfile('randomForest') or not os.path.isfile('Tree'):
thread1 = Thread(target=LogReg.train())
thread2 = Thread(target=SVMFIle.train())
thread3 = Thread(target=TreeClass.train())
thread4 = Thread(target=RandomForestFile.train())
thread1.start()
thread2.start()
thread3.start()
thread4.start()
thread1.join()
thread2.join()
thread3.join()
thread4.join()
app = QtWidgets.QApplication(sys.argv)
w = Ui()
w.show()
sys.exit(app.exec_())
### Tree
TTree = {}
for line in open("../Data/nodes.dmp"):
line = [i.strip() for i in line.split("|")][0:3]
Node = TaxonTerm(line[0], "", line[2])
Node.Parent = line[1]
Node.Childs = set()
Node.seqs = []
TTree[line[0]] = Node
for ID, Node in TTree.items():
Node.Parent = TTree[Node.Parent]
Node.Parent.Childs.add(Node)
# print len(TTree)
### genus Data
Data = pd.read_excel("../Data/ZYName.xlsx", 0, index_col=0)
for i in Data.index:
print i, Data.loc[i, "Name"]
IDs = set()
IDs = tc.getAllNodeofGenus(TTree[str(Data.loc[i, "Tid"])], IDs)
out = []
for ID in IDs:
out.append([ID, ID2Name[ID], TTree[ID].Rank])
# print out
out = pd.DataFrame(np.array(out), columns=["ID", "Name", "Rank"])
out.to_excel(
"../Result0923/Sepcies/" + str(i) + "_" + str(Data.loc[i, "Tid"]) + "_" + str(Data.loc[i, "TName"]) + ".xls"
)
print "Done!"