Update app.py

app.py

@@ -121,13 +121,10 @@ def predict(file_obj):
     try:
         sequences = parse_fasta(text)
         for header, seq in sequences:
-            # Get k-mer vector
-            kmer_vector = sequence_to_kmer_vector(seq)
-            kmer_vector = scaler.transform(kmer_vector.reshape(1, -1))
-            X_tensor = torch.FloatTensor(kmer_vector).to(device)
-            
-            # Get raw frequency vector before scaling
+            # Get raw frequency vector and scaled vector
             raw_freq_vector = sequence_to_kmer_vector(seq)
+            kmer_vector = scaler.transform(raw_freq_vector.reshape(1, -1))
+            X_tensor = torch.FloatTensor(kmer_vector).to(device)
             
             # Get predictions and feature importance
             with torch.no_grad():
@@ -138,17 +135,18 @@ def predict(file_obj):
             importance = model.get_feature_importance(X_tensor)
             kmer_importance = importance[0].cpu().numpy()
             
-            # Normalize importance scores to match original scale
+            # Normalize importance scores to original scale
             kmer_importance = kmer_importance / np.max(np.abs(kmer_importance)) * 0.002
             
-            # Get top 10 k-mers
+            # Get top 10 k-mers based on absolute importance
             top_k = 10
             top_indices = np.argsort(np.abs(kmer_importance))[-top_k:][::-1]
             important_kmers = [
                 {
                     'kmer': list(kmer_dict.keys())[list(kmer_dict.values()).index(i)],
                     'importance': float(kmer_importance[i]),
-                    'frequency': float(raw_freq_vector[i])
+                    'frequency': float(raw_freq_vector[i]),
+                    'scaled': float(kmer_vector[0][i])
                 }
                 for i in top_indices
             ]
@@ -166,7 +164,7 @@ Non-human probability: {float(probs[0][0]):0.4f}
 Most influential k-mers:"""
             
             for kmer in important_kmers:
-                result += f"\n  {kmer['kmer']}: importance={kmer['importance']:.4f}, raw_freq={kmer['raw_freq']:.4f} ({kmer['raw_freq']*100:.2f}%), scaled_freq={kmer['scaled_freq']:.4f}"
+                result += f"\n  {kmer['kmer']}: importance={kmer['importance']:.4f}, frequency={kmer['frequency']:.4f} ({kmer['frequency']*100:.2f}%), scaled={kmer['scaled']:.4f}"
             
             results.append(result)
     except Exception as e: