class DNASequenceClassifier:
def __init__(self, sequence_length=200, num_classes=2):
self.sequence_length = sequence_length
self.num_classes = num_classes
self.mannequin = None
self.historical past = None
def one_hot_encode(self, sequences):
mapping = {'A': 0, 'T': 1, 'G': 2, 'C': 3}
encoded = np.zeros((len(sequences), self.sequence_length, 4))
for i, seq in enumerate(sequences):
for j, nucleotide in enumerate(seq[:self.sequence_length]):
if nucleotide in mapping:
encoded[i, j, mapping[nucleotide]] = 1
return encoded
def attention_layer(self, inputs, title="consideration"):
attention_weights = layers.Dense(1, activation='tanh', title=f"{title}_weights")(inputs)
attention_weights = layers.Flatten()(attention_weights)
attention_weights = layers.Activation('softmax', title=f"{title}_softmax")(attention_weights)
attention_weights = layers.RepeatVector(inputs.form[-1])(attention_weights)
attention_weights = layers.Permute([2, 1])(attention_weights)
attended = layers.Multiply(title=f"{title}_multiply")([inputs, attention_weights])
return layers.GlobalMaxPooling1D()(attended)
def build_model(self):
inputs = layers.Enter(form=(self.sequence_length, 4), title="dna_input")
conv_layers = []
filter_sizes = [3, 7, 15, 25]
for i, filter_size in enumerate(filter_sizes):
conv = layers.Conv1D(
filters=64,
kernel_size=filter_size,
activation='relu',
padding='identical',
title=f"conv_{filter_size}"
)(inputs)
conv = layers.BatchNormalization(title=f"bn_conv_{filter_size}")(conv)
conv = layers.Dropout(0.2, title=f"dropout_conv_{filter_size}")(conv)
attended = self.attention_layer(conv, title=f"attention_{filter_size}")
conv_layers.append(attended)
if len(conv_layers) > 1:
merged = layers.Concatenate(title="concat_multiscale")(conv_layers)
else:
merged = conv_layers[0]
dense = layers.Dense(256, activation='relu', title="dense_1")(merged)
dense = layers.BatchNormalization(title="bn_dense_1")(dense)
dense = layers.Dropout(0.5, title="dropout_dense_1")(dense)
dense = layers.Dense(128, activation='relu', title="dense_2")(dense)
dense = layers.BatchNormalization(title="bn_dense_2")(dense)
dense = layers.Dropout(0.3, title="dropout_dense_2")(dense)
if self.num_classes == 2:
outputs = layers.Dense(1, activation='sigmoid', title="output")(dense)
loss="binary_crossentropy"
metrics = ['accuracy', 'precision', 'recall']
else:
outputs = layers.Dense(self.num_classes, activation='softmax', title="output")(dense)
loss="categorical_crossentropy"
metrics = ['accuracy']
self.mannequin = keras.Mannequin(inputs=inputs, outputs=outputs, title="DNA_CNN_Classifier")
optimizer = keras.optimizers.Adam(
learning_rate=0.001,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-7
)
self.mannequin.compile(
optimizer=optimizer,
loss=loss,
metrics=metrics
)
return self.mannequin
def generate_synthetic_data(self, n_samples=10000):
sequences = []
labels = []
positive_motifs = ['TATAAA', 'CAAT', 'GGGCGG', 'TTGACA']
negative_motifs = ['AAAAAAA', 'TTTTTTT', 'CCCCCCC', 'GGGGGGG']
nucleotides = ['A', 'T', 'G', 'C']
for i in vary(n_samples):
sequence="".be a part of(random.decisions(nucleotides, okay=self.sequence_length))
if i < n_samples // 2:
motif = random.selection(positive_motifs)
pos = random.randint(0, self.sequence_length - len(motif))
sequence = sequence[:pos] + motif + sequence[pos + len(motif):]
label = 1
else:
if random.random() < 0.3:
motif = random.selection(negative_motifs)
pos = random.randint(0, self.sequence_length - len(motif))
sequence = sequence[:pos] + motif + sequence[pos + len(motif):]
label = 0
sequences.append(sequence)
labels.append(label)
return sequences, np.array(labels)
def prepare(self, X_train, y_train, X_val, y_val, epochs=50, batch_size=32):
callbacks = [
keras.callbacks.EarlyStopping(
monitor="val_loss",
patience=10,
restore_best_weights=True
),
keras.callbacks.ReduceLROnPlateau(
monitor="val_loss",
factor=0.5,
patience=5,
min_lr=1e-6
)
]
self.historical past = self.mannequin.match(
X_train, y_train,
validation_data=(X_val, y_val),
epochs=epochs,
batch_size=batch_size,
callbacks=callbacks,
verbose=1
)
return self.historical past
def evaluate_and_visualize(self, X_test, y_test):
y_pred_proba = self.mannequin.predict(X_test)
y_pred = (y_pred_proba > 0.5).astype(int).flatten()
print("Classification Report:")
print(classification_report(y_test, y_pred))
fig, axes = plt.subplots(2, 2, figsize=(15, 10))
axes[0,0].plot(self.historical past.historical past['loss'], label="Coaching Loss")
axes[0,0].plot(self.historical past.historical past['val_loss'], label="Validation Loss")
axes[0,0].set_title('Coaching Historical past - Loss')
axes[0,0].set_xlabel('Epoch')
axes[0,0].set_ylabel('Loss')
axes[0,0].legend()
axes[0,1].plot(self.historical past.historical past['accuracy'], label="Coaching Accuracy")
axes[0,1].plot(self.historical past.historical past['val_accuracy'], label="Validation Accuracy")
axes[0,1].set_title('Coaching Historical past - Accuracy')
axes[0,1].set_xlabel('Epoch')
axes[0,1].set_ylabel('Accuracy')
axes[0,1].legend()
cm = confusion_matrix(y_test, y_pred)
sns.heatmap(cm, annot=True, fmt="d", ax=axes[1,0], cmap='Blues')
axes[1,0].set_title('Confusion Matrix')
axes[1,0].set_ylabel('Precise')
axes[1,0].set_xlabel('Predicted')
axes[1,1].hist(y_pred_proba[y_test==0], bins=50, alpha=0.7, label="Destructive", density=True)
axes[1,1].hist(y_pred_proba[y_test==1], bins=50, alpha=0.7, label="Optimistic", density=True)
axes[1,1].set_title('Prediction Rating Distribution')
axes[1,1].set_xlabel('Prediction Rating')
axes[1,1].set_ylabel('Density')
axes[1,1].legend()
plt.tight_layout()
plt.present()
return y_pred, y_pred_proba