XspecT 0.1.2__py3-none-any.whl → 0.2.0__py3-none-any.whl

xspect/Classifier.py

@@ -1,142 +0,0 @@
-from sklearn.svm import SVC
-import csv
-from copy import deepcopy
-
-
-def cut_csv(csv_file, lst, table=False):
-    """Returns desired data from Training_data"""
-    r = csv.reader(open(csv_file))
-    m = list(r)
-    selected = deepcopy(lst)
-
-    header = m[0]
-    m = m[1:]
-    labels = header[1:-1]
-
-    X_train = []
-    y_train = []
-    files = []
-
-    if selected[8] and len(header[8:-1]) > 0:
-        # Added Genomes selected
-        del selected[8]
-        selected = selected + ([True] * len(header[9:-1]))
-
-    else:
-        # Added Genomes not selected
-        del selected[8]
-        selected = selected + ([False] * len(header[9:-1]))
-
-    # creating matrix
-    for i in range(len(m)):
-        X_train.append(m[i][1:-1])
-        y_train.append(m[i][-1])
-        files.append(m[i][0])
-
-    # Deleting Cols
-    for i in range(len(X_train)):
-        for j in range(len(X_train[i]) - 1, -1, -1):
-            if selected[j]:
-                pass
-            else:
-                del X_train[i][j]
-
-    # Deleting Rows
-    valid = ["None"]
-    for i in range(len(selected)):
-        if selected[i]:
-            valid.append(labels[i])
-
-    for i in range(len(X_train) - 1, -1, -1):
-        if y_train[i] not in valid:
-            del y_train[i]
-            del X_train[i]
-            del files[i]
-
-    if table:
-        # Inserting Infos for Table
-        for i in range(len(X_train)):
-            X_train[i].insert(0, files[i])
-            X_train[i].append(y_train[i])
-
-        for i in range(len(header) - 1, -1, -1):
-            if header[i] not in valid:
-                del header[i]
-
-        header.insert(0, "File")
-        header.append("Label")
-
-        X_train.insert(0, header)
-
-    else:
-        pass
-
-    return X_train, y_train
-
-
-def cut_csv_spec(csv_file):
-    """Returns svm Training_data"""
-    # read the training-data
-    r = csv.reader(open(csv_file))
-    m = list(r)
-    header = m[0]
-    m = m[1:]
-    X_train = []
-    y_train = []
-
-    # creating matrix as input for the classifier
-    for i in range(len(m)):
-        X_train.append(m[i][1:-1])
-        y_train.append(m[i][-1])
-
-    return X_train, y_train
-
-
-def classify(csv_file, result, lst):
-    """Classifys Result-vector and calculates needed vectors"""
-    r = csv.reader(open(csv_file))
-    m = list(r)
-    # deciding which kernel-function will be used
-    if m[0][1] == "IC1":
-        mode = "ClAssT"
-        X_train, y_train = cut_csv(csv_file, lst)
-        svm = SVC(kernel="poly", C=1.0).fit(X_train, y_train)
-    else:
-        mode = "XspecT"
-        X_train, y_train = cut_csv_spec(csv_file)
-        svm = SVC(kernel="rbf", C=1.5).fit(X_train, y_train)
-    # perform a prediction using the svm
-    prediction = svm.predict([result])
-    if mode == "XspecT":
-        if max(result) < 0.3:
-            prediction = ["sp.", 0]
-    else:
-        if max(result) < 0.3:
-            prediction = ["None", 0]
-
-    return prediction[0]
-
-
-def IC3_classify(result_2):
-    ic = "International Clonetype 3 (ST32 or ST250)"
-    m_3 = [
-        ["GCF_000278625.1", 1.0, ic],
-        ["GCF_001674185.1", 0.86, ic],
-        ["fictional", 0.85, "NONE of the selected Clonetypes or Genomes"],
-        ["fictional", 0.01, "NONE of the selected Clonetypes or Genomes"],
-    ]
-
-    X = []
-    y = []
-    for i in range(len(m_3)):
-        X.append(m_3[i][1])
-        y.append(m_3[i][2])
-
-    for i in range(len(X)):
-        X[i] = [X[i]]
-    svm_IC3 = SVC(kernel="poly", C=1).fit(X, y)
-
-    return svm_IC3.predict([result_2]), result_2[0]
-
-
-# https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html

xspect/OXA_Table.py

@@ -1,53 +0,0 @@
-import json
-from Bio import SeqIO
-import os
-
-
-class OXATable:
-    def __init__(self):
-        self.kmere = {}
-        self.total = 0
-        self.found = 0
-
-    def create_table(self, directory):
-        """Reads in fasta file, creates Dictionary with k-mer counter"""
-        # taking all fasta files for one big table
-        oxas = {}
-        files = os.listdir(directory)
-        for i in range(len(files)):
-            kmere = {}
-            file = directory + files[i]
-            for sequence in SeqIO.parse(file, "fasta"):
-                for j in range(0, len(sequence.seq) - 20 + 1):
-                    kmer = str(sequence.seq[j : j + 20])
-                    count = kmere.get(kmer, 0)
-                    kmere[kmer] = count + 1
-            oxas[files[i][:-6]] = kmere
-
-    def lookup(self, gene, kmer):
-        """Tests if kmer in dictionary, if so: reduces the counter"""
-
-        # Only returns True if kmer has been found and there was one left
-        if kmer in self.kmere[gene]:
-            if self.kmere[gene][kmer] > 0:
-                self.kmere[gene][kmer] -= 1
-                return True
-            else:
-                return False
-        else:
-            return False
-
-    def save_dic(self, path):
-        """writes dictionary to file using json"""
-        json.dump(self.kmere, open(path, "w"))
-
-    def read_dic(self, path):
-        """Reads dictionary from file using json"""
-        self.kmere = json.load(open(path))
-
-    def cleanup(self):
-        self.kmere = {}
-
-    def get_counter(self, path=r"filter/OXAs_dict/counter.txt"):
-        counter = json.load(open(path))
-        return counter