Added duplicate deletion, limit to input SMILES and updated infos

README.md

@@ -63,7 +63,9 @@ C1CCCCC1O
 C1=CC=C(C=C1)NC(=O)C2=CC=CC=N2
 ```
 
-The model will predict binding affinity for each drug-target pair sequentially. For visualizations, the results will display only the last SMILES entry.
+The model processes each drug–target pair in sequence to predict binding affinity. For visualization, only the final SMILES entry is shown.
+If “Remove duplicate SMILES” is enabled, any repeated SMILES strings are filtered out before analysis.
+Final results follow the original SMILES order, with adjusted index numbers when duplicates have been removed.
 
 ### Programmatic usage
 

app.py

@@ -334,19 +334,41 @@ class DrugTargetInteractionApp:
 # Initialize the app
 app = DrugTargetInteractionApp()
 
-def predict_wrapper(target_seq, drug_smiles):
+def smiles_preprocessing(drug_smiles):
+    drugs = drug_smiles.strip().split("\n")
+    
+    # Remove molecule duplicates in O(n) while preserving the order
+    seen = set()
+    sorted_drugs = []
+
+    kept = 0
+    
+    for x in drugs:
+        if x not in seen:
+            seen.add(x)
+            sorted_drugs.append(x)
+            kept += 1
+
+    logger.info(f"{kept-len(drugs)} duplicate smiles removed!")
+    return sorted_drugs    
+
+def predict_wrapper(target_seq, drug_smiles, remove_dups):
     """Wrapper function for Gradio interface"""
     if not target_seq.strip() or not drug_smiles.strip():
         return "Please provide both target sequence and drug SMILES."
     
-    return app.predict_interaction(target_seq, drug_smiles.split("\n"))
+    target_seq = target_seq.strip() 
+    drug_smiles = smiles_preprocessing(drug_smiles, remove_dups)
+    return app.predict_interaction(target_seq, drug_smiles)
 
-def visualize_wrapper(target_seq, drug_smiles):
+def visualize_wrapper(target_seq, drug_smiles, remove_dups):
     """Wrapper function for visualization"""
     if not target_seq.strip() or not drug_smiles.strip():
         return None, None, None, "Please provide both target sequence and drug SMILES."
 
-    return app.visualize_interaction(target_seq, drug_smiles.split("\n"))
+    target_seq = target_seq.strip() 
+    drug_smiles = smiles_preprocessing(drug_smiles, remove_dups)
+    return app.visualize_interaction(target_seq, drug_smiles)
 
 def load_model_wrapper(model_path):
     """Wrapper function to load model"""
@@ -389,9 +411,15 @@ with gr.Blocks(title="Drug-Target Interaction Predictor", theme=gr.themes.Soft()
                         "For multiple SMILES, enter each on a new line:\n"
                         "CC(C)CC1=CC=C(C=C1)C(C)C(=O)O\n"
                         "C1CCCCC1O",
-                    lines=3
+                    lines=4,
+                    max_lines=2000
                 )
                 
+                remove_dups_checkbox = gr.Checkbox(
+                    label="Remove duplicate SMILES",
+                    value=True
+                )  
+
                 with gr.Row():
                     predict_btn = gr.Button("🚀 Predict Interaction", variant="primary", size="lg")
                     visualize_btn = gr.Button("📊 Generate Visualizations", variant="secondary", size="lg")
@@ -417,7 +445,7 @@ with gr.Blocks(title="Drug-Target Interaction Predictor", theme=gr.themes.Soft()
                     "C1=CC=C(C=C1)NC(=O)C2=CC=CC=N2"
                 ]
             ],
-            inputs=[target_input, drug_input],
+            inputs=[target_input, drug_input, remove_dups_checkbox],
             outputs=prediction_output,
             fn=predict_wrapper,
             cache_examples=False
@@ -426,13 +454,14 @@ with gr.Blocks(title="Drug-Target Interaction Predictor", theme=gr.themes.Soft()
         # Button click events
         predict_btn.click(
             fn=predict_wrapper,
-            inputs=[target_input, drug_input],
+            inputs=[target_input, drug_input, remove_dups_checkbox],
             outputs=prediction_output
         )
+
         
-        def visualize_and_update(target_seq, drug_smiles):
+        def visualize_and_update(target_seq, drug_smiles, remove_dups):
             """Generate visualizations and update both status and state"""
-            img1, img2, img3, status = visualize_wrapper(target_seq, drug_smiles)
+            img1, img2, img3, status = visualize_wrapper(target_seq, drug_smiles, remove_dups)
             # Combine prediction result with visualization status
             combined_status = status + "\n\nVisualization analysis complete. Please navigate to the Visualizations tab to view the generated images."
             if len(drug_smiles) > 1:
@@ -442,11 +471,11 @@ with gr.Blocks(title="Drug-Target Interaction Predictor", theme=gr.themes.Soft()
         
         visualize_btn.click(
             fn=visualize_and_update,
-            inputs=[target_input, drug_input],
+            inputs=[target_input, drug_input, remove_dups_checkbox],
             outputs=[viz_state1, viz_state2, viz_state3, prediction_output],
-            api_name="visualize_and_update"  # Make this API accessible
+            api_name="visualize_and_update"
         )
-    
+
     with gr.Tab("📊 Visualizations"):
         gr.HTML("""
         <div style="text-align: center; margin-bottom: 20px;">
@@ -588,33 +617,34 @@ with gr.Blocks(title="Drug-Target Interaction Predictor", theme=gr.themes.Soft()
     with gr.Tab("ℹ️ About"):
         gr.Markdown("""
         ## About this application
-        
-        This application implements DLRNA-BERTa, a Dual Language RoBERTa Transformer model for predicting drug to RNA target interactions. The model architecture includes:
-        
-        - **Target encoder**: Processes RNA sequences using RNA-BERTa
-        - **Drug encoder**: Processes molecular SMILES notation using ChemBERTa
-        - **Cross-attention mechanism**: Captures interactions between drugs and targets
-        - **Regression head**: Predicts binding affinity scores (pKd values)
-        
-        ### Input requirements:
-        - **Target sequence**: RNA sequence of the target (nucleotide sequences: A, U, G, C)
-        - **Drug SMILES**: Simplified Molecular Input Line Entry System notation
-          - **Batch prediction**: Enter multiple SMILES strings, one per line, to predict binding affinity for multiple drugs against the same target
-        
-        ### Model features:
-        - Cross-attention for drug-target interaction modeling
-        - Dropout for regularization
-        - Layer normalization for stable training
-        - Interpretability mode for contribution and attention visualization
-        - Support for batch predictions with multiple SMILES entries
-        
-        ### Usage tips:
-        1. Load a trained model using the Model Settings tab (optional)
-        2. Enter a RNA sequence and drug SMILES in the Prediction & Analysis tab
-        3. For batch predictions, enter multiple SMILES strings (one per line) in the drug SMILES field
-        4. Click "Predict Interaction" for binding affinity prediction only
-        5. Click "Generate Visualizations" to create detailed interaction analysis - results will appear in the Visualizations tab
-        6. Note: Visualizations are generated only for the last SMILES entry when using batch mode
+
+        This application implements DLRNA-BERTa, a Dual Language RoBERTa Transformer model for predicting drug-to-RNA target interactions. The architecture combines:
+
+        - **Target encoder**: RNA-BERTa for processing RNA sequences  
+        - **Drug encoder**: ChemBERTa for SMILES representation  
+        - **Cross-attention mechanism**: Captures interactions between drug and target  
+        - **Regression head**: Predicts binding affinity (pKd)
+
+        ### Input requirements
+        - **Target sequence**: RNA sequence (A, U, G, C)
+        - **Drug SMILES**: One or more SMILES strings  
+        - For batch mode, enter each SMILES on a new line (max 2000 entries)
+        - A checkbox option allows automatic removal of duplicate SMILES before prediction
+
+        ### Model features
+        - Cross-attention for drug-target interaction modeling  
+        - Regularization via dropout  
+        - Layer normalization for stable training  
+        - Dedicated interpretability mode for visualization  
+        - Batch prediction with optional de-duplication  
+
+        ### Usage tips
+        1. Load a model (optional) in the Model Settings tab  
+        2. Enter an RNA sequence and one or more SMILES strings  
+        3. Use the **“Remove duplicate SMILES”** checkbox if you want duplicates filtered automatically  
+        4. Click *Predict Interaction* for affinity scores  
+        5. Click *Generate Visualizations* for interpretability plots  
+        6. Visualizations are produced only for the final SMILES entry in batch mode  
         
         For best results, ensure your input sequences are properly formatted and within reasonable length limits (max 512 tokens).