Update app.py

app.py

@@ -46,40 +46,25 @@ selected_features = [
 ]
 
 def extract_features(sequence):
-    """Extracts features, aligns, and normalizes, prioritizing AADipeptide."""
-    try:
-        # 1. Calculate Dipeptide Composition (as per your request)
-        dipeptide_features = AAComposition.CalculateAADipeptideComposition(sequence)
-        dipeptide_values = list(dipeptide_features.values())
-        dipeptide_array = np.array(dipeptide_values).reshape(1, -1) #Correct shape
+ 
+    dipeptide_features = AAComposition.CalculateAADipeptideComposition(sequence)
 
-        # 2. Calculate other features
-        auto_features = Autocorrelation.CalculateAutoTotal(sequence)
-        ctd_features = CTD.CalculateCTD(sequence)
-        pseudo_features = PseudoAAC.GetAPseudoAAC(sequence)
-        all_features = {**auto_features, **ctd_features, **pseudo_features,**dipeptide_features}
+    # 2. Calculate other features
+    auto_features = Autocorrelation.CalculateAutoTotal(sequence)
+    ctd_features = CTD.CalculateCTD(sequence)
+    pseudo_features = PseudoAAC.GetAPseudoAAC(sequence)
+    all_features = {**auto_features, **ctd_features, **pseudo_features,**dipeptide_features}
+    all_features = list(all_features.values())
+    all_features = np.array(all_features).reshape(-1, 1) #Correct shape
+    normalized_features = scaler.transform(all_features.T)
+    normalized_features = normalized_features.flatten()  
 
-         # Create a DataFrame for ALL features
-        all_features_df = pd.DataFrame([all_features])
+    selected_feature_dict = {feature: normalized_features[i] for i, feature in enumerate(selected_features) if feature in all_features}
+    selected_feature_df = pd.DataFrame([selected_feature_dict])
+    selected_feature_array = selected_feature_df.T.to_numpy()
 
-        # --- Feature Selection and Alignment ---
-        present_features = [col for col in selected_features if col in all_features_df.columns]
-        selected_df = all_features_df[present_features]
-        aligned_df = pd.DataFrame(columns=selected_features)
-        aligned_df.update(selected_df)
-        aligned_df = aligned_df.fillna(0)
-
-
-        # Normalize
-        normalized_features = scaler.transform(aligned_df)
-        return normalized_features
-
-    except (ZeroDivisionError, KeyError, TypeError, ValueError) as e:
-        print(f"Error during feature extraction: {e}")
-        return None
-    except Exception as e:
-         print(f"An unexpected error occurred: {e}")
-         return None
+    return selected_feature_array
+    
 
 
 def predict(sequence):