get-hga-training-test-py27.py
8.18 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
# Get training and test data set for deep learning from sequence data set
# obtained from FASTA and HGA data sets (see script get-hga-sequences-py3.py)
# Input tab-separated format:
# Sequences: hga-sequences-toy.txt
# Output one-hot encoding format:
# Each sequence as a one-hot encoding WHAT array or matrix
# Run:
# python3 get-hga-training-test-py27.py
# --inputFile hga-sequences-toy.txt
# --inputPath /home/cmendezc/gitlab-deep-learning-workshop/data-sets/human-genome-annotation
# --outputTraining hga-sequences-training.txt
# --outputTest hga-sequences-test.txt
# --outputPath /home/cmendezc/gitlab-deep-learning-workshop/data-sets/human-genome-annotation
# python get-hga-training-test-py3.py --inputFile hga-sequences-1000.txt --inputPath /home/cmendezc/gitlab-deep-learning-workshop/data-sets/human-genome-annotation --outputTraining hga-sequences-training.txt --outputTest hga-sequences-test.txt --outputPath /home/cmendezc/gitlab-deep-learning-workshop/data-sets/human-genome-annotation
# LAVIS
# qlogin
# python get-hga-training-test-py27.py
# --inputFile hga-sequences-1000.txt
# --inputPath /mnt/Genoma/amedina/cmendez/gitlab-deep-learning-workshop/data-sets/human-genome-annotation
# --outputTraining hga-sequences-training.txt
# --outputTest hga-sequences-test.txt
# --outputPath /mnt/Genoma/amedina/cmendez/gitlab-deep-learning-workshop/data-sets/human-genome-annotation
# python get-hga-training-test-py27.py --inputFile hga-sequences-toy.txt --inputPath /mnt/Genoma/amedina/cmendez/gitlab-deep-learning-workshop/data-sets/human-genome-annotation --outputTraining hga-sequences-training.txt --outputTest hga-sequences-test.txt --outputPath /mnt/Genoma/amedina/cmendez/gitlab-deep-learning-workshop/data-sets/human-genome-annotation
import argparse
import pandas as pd
import os
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
import numpy as np
from sklearn.model_selection import train_test_split
from tensorflow.keras.layers import Conv1D, Dense, MaxPooling1D, Flatten
from tensorflow.keras.models import Sequential
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix
import itertools
if __name__ == "__main__":
parser = argparse.ArgumentParser(description='Get training and test data sets for Human Genome Annotation.')
parser.add_argument('--inputFile', dest='inputFile',
help='Input file')
parser.add_argument('--inputPath', dest='inputPath',
help='Input path')
parser.add_argument('--outputTraining', dest='outputTraining',
help='Output training file')
parser.add_argument('--outputValidation', dest='outputValidation',
help='Output training file')
parser.add_argument('--outputTest', dest='outputTest',
help='Output test file')
parser.add_argument('--outputPath', dest='outputPath',
help='Output path for training, validation, and testing')
args = parser.parse_args()
# To one-hot encoding taken from: https://colab.research.google.com/drive/17E4h5aAOioh5DiTo7MZg4hpL6Z_0FyWr#scrollTo=IPJD6PuDnaS6
# The LabelEncoder encodes a sequence of bases as a sequence of integers.
integer_encoder = LabelEncoder()
# The OneHotEncoder converts an array of integers to a sparse matrix where
# each row corresponds to one possible value of each feature.
one_hot_encoder = OneHotEncoder(categories='auto')
input_features = []
# Read file with sequences
with open(os.path.join(args.inputPath, args.inputFile), mode="r") as tabfile:
df = pd.read_csv(tabfile, delimiter='\t')
print("All rows in df: {}".format(len(df.index)))
df_filtered = df.loc[(df['label'] == "exon") | (df['label'] == "utr")]
print("Only exon and utr rows in df: {}".format(len(df_filtered.index)))
# print("df: {}".format(df))
sequences = df_filtered['sequence']
labels = df_filtered['label']
max_exon_length = 0
max_utr_length = 0
# Getting the max length of sequences
for sequence, label in zip(sequences, labels):
if label == "exon":
if len(sequence) > max_exon_length:
max_exon_length = len(sequence)
elif label == "utr":
if len(sequence) > max_utr_length:
max_utr_length = len(sequence)
print("Max exon length: {}".format(max_exon_length))
print("Max utr length: {}".format(max_utr_length))
# Fill sequence with X char to get max length
# One-hot-encoding of sequences
for sequence, label in zip(sequences, labels):
if label == "exon":
if len(sequence) < max_exon_length:
sequence_adjust = sequence.ljust(max_exon_length + len(sequence), 'X')
elif label == "utr":
if len(sequence) < max_utr_length:
sequence_adjust = sequence.ljust(max_utr_length + len(sequence), 'X')
print("Length sequence_adjust: {}".format(len(sequence_adjust)))
integer_encoded = integer_encoder.fit_transform(list(sequence_adjust))
integer_encoded = np.array(integer_encoded).reshape(-1, 1)
one_hot_encoded = one_hot_encoder.fit_transform(integer_encoded)
input_features.append(one_hot_encoded.toarray())
# Print first sequence and one-hot-encoding
np.set_printoptions(threshold=40)
input_features = np.stack(input_features)
print("Example sequence\n-----------------------")
print('DNA Sequence #1:\n', sequences[0][:10], '...', sequences[0][-10:])
print('One hot encoding of Sequence #1:\n', input_features[0].T)
# One-hot-encoding of labels
one_hot_encoder = OneHotEncoder(categories='auto')
labels = np.array(labels).reshape(-1, 1)
input_labels = one_hot_encoder.fit_transform(labels).toarray()
# Print labels and one-hot-encoding
print('Labels:\n', labels.T)
print('One-hot encoded labels:\n', input_labels.T)
# Split one-hot-encoding data into training, and test data sets
train_features, test_features, train_labels, test_labels = train_test_split(
input_features, input_labels, test_size=0.25, random_state=42)
# Model definition
model = Sequential()
model.add(Conv1D(filters=32, kernel_size=12,
input_shape=(train_features.shape[1], 4)))
model.add(MaxPooling1D(pool_size=4))
model.add(Flatten())
model.add(Dense(16, activation='relu'))
model.add(Dense(2, activation='softmax'))
model.compile(loss='binary_crossentropy', optimizer='adam',
metrics=['binary_accuracy'])
model.summary()
# Model training and validation
history = model.fit(train_features, train_labels,
epochs=50, verbose=0, validation_split=0.25)
# Plot training-validation loss
plt.figure()
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'validation'])
# plt.show()
plt.savefig('training-validation-loss.png')
# Plot training-validation accuracy
plt.figure()
plt.plot(history.history['binary_accuracy'])
plt.plot(history.history['val_binary_accuracy'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'validation'])
# plt.show()
plt.savefig('training-validation-binary-accuracy.png')
# Predict with rest data set
predicted_labels = model.predict(np.stack(test_features))
# Print confusion matrix
cm = confusion_matrix(np.argmax(test_labels, axis=1),
np.argmax(predicted_labels, axis=1))
print('Confusion matrix:\n', cm)
cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]
# Plot confusion matrix
plt.imshow(cm, cmap=plt.cm.Blues)
plt.title('Normalized confusion matrix')
plt.colorbar()
plt.xlabel('True label')
plt.ylabel('Predicted label')
plt.xticks([0, 1]);
plt.yticks([0, 1])
plt.grid('off')
for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
plt.text(j, i, format(cm[i, j], '.2f'),
horizontalalignment='center',
color='white' if cm[i, j] > 0.5 else 'black')
plt.savefig('training-validation-confusion-matrix.png')