Update app.py

app.py

@@ -1,46 +1,52 @@
 import gradio as gr
 import requests
-from rdkit import Chem
-from rdkit.Chem import Descriptors
-import pandas as pd
+import json
 from inference import DDIPredictor
-import torch
 import re
 
 # Initialize your trained model
 predictor = DDIPredictor(".")
 
 def fetch_pubchem_data(drug_name):
-    """Fetch drug data from PubChem by name"""
+    """Fetch comprehensive drug data from PubChem by name"""
     try:
         # First, search for the compound ID
         search_url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/name/{drug_name}/cids/JSON"
-        search_response = requests.get(search_url)
+        search_response = requests.get(search_url, timeout=10)
         
         if search_response.status_code != 200:
             return None, f"Drug '{drug_name}' not found in PubChem"
         
         cid = search_response.json()['IdentifierList']['CID'][0]
         
-        # Fetch compound data
-        compound_url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/{cid}/property/CanonicalSMILES,MolecularWeight,XLogP,TPSA/JSON"
-        compound_response = requests.get(compound_url)
+        # Fetch comprehensive compound data
+        properties = [
+            'CanonicalSMILES', 'MolecularWeight', 'XLogP', 'TPSA',
+            'RotatableBondCount', 'HBondDonorCount', 'HBondAcceptorCount',
+            'Complexity', 'Charge', 'ExactMass'
+        ]
+        
+        compound_url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/{cid}/property/{','.join(properties)}/JSON"
+        compound_response = requests.get(compound_url, timeout=10)
         
         if compound_response.status_code != 200:
-            return None, "Failed to fetch compound data"
+            return None, "Failed to fetch compound properties"
         
         data = compound_response.json()['PropertyTable']['Properties'][0]
+        data['CID'] = cid
         
-        # Calculate additional properties from SMILES
-        mol = Chem.MolFromSmiles(data['CanonicalSMILES'])
-        if mol:
-            data['RotatableBondCount'] = Descriptors.NumRotatableBonds(mol)
-            data['HBondDonorCount'] = Descriptors.NumHDonors(mol)
-            data['HBondAcceptorCount'] = Descriptors.NumHAcceptors(mol)
-            data['Complexity'] = Descriptors.MolWt(mol)  # Simplified complexity
+        # Get IUPAC name for better identification
+        iupac_url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/{cid}/property/IUPACName/JSON"
+        iupac_response = requests.get(iupac_url, timeout=10)
+        if iupac_response.status_code == 200:
+            data['IUPACName'] = iupac_response.json()['PropertyTable']['Properties'][0]['IUPACName']
+        else:
+            data['IUPACName'] = drug_name
         
         return data, None
         
+    except requests.exceptions.Timeout:
+        return None, "PubChem API timeout - please try again"
     except Exception as e:
         return None, f"Error fetching data: {str(e)}"
 
@@ -49,64 +55,101 @@ def generate_interaction_description(drug1_data, drug2_data):
     
     descriptions = []
     
-    # Based on molecular weight difference
-    mw_diff = abs(drug1_data['MolecularWeight'] - drug2_data['MolecularWeight'])
+    # Molecular weight analysis
+    mw1 = drug1_data.get('MolecularWeight', 0)
+    mw2 = drug2_data.get('MolecularWeight', 0)
+    mw_diff = abs(mw1 - mw2)
+    
     if mw_diff > 500:
-        descriptions.append("Significant molecular size difference may affect metabolism")
+        descriptions.append("Significant molecular size difference may affect metabolic pathways")
+    elif mw_diff > 200:
+        descriptions.append("Moderate molecular size difference could influence distribution")
+    
+    # Lipophilicity analysis (XLogP)
+    logp1 = drug1_data.get('XLogP', 0)
+    logp2 = drug2_data.get('XLogP', 0)
+    logp_diff = abs(logp1 - logp2)
     
-    # Based on lipophilicity (XLogP)
-    logp_diff = abs(drug1_data.get('XLogP', 0) - drug2_data.get('XLogP', 0))
     if logp_diff > 3:
-        descriptions.append("Differing lipophilicity may influence distribution and clearance")
+        descriptions.append("Differing lipophilicity may affect membrane permeability and clearance rates")
+    elif logp_diff > 1.5:
+        descriptions.append("Variation in lipophilicity could influence tissue distribution")
+    
+    # Polar surface area analysis
+    tpsa1 = drug1_data.get('TPSA', 0)
+    tpsa2 = drug2_data.get('TPSA', 0)
+    tpsa_diff = abs(tpsa1 - tpsa2)
     
-    # Based on TPSA (polar surface area)
-    tpsa_diff = abs(drug1_data.get('TPSA', 0) - drug2_data.get('TPSA', 0))
     if tpsa_diff > 100:
-        descriptions.append("Varying polar surface areas suggest different membrane permeability")
+        descriptions.append("Significant difference in polar surface area suggests varying bioavailability")
+    elif tpsa_diff > 50:
+        descriptions.append("Moderate difference in polar surface area may affect absorption")
+    
+    # Hydrogen bonding analysis
+    h_donor1 = drug1_data.get('HBondDonorCount', 0)
+    h_donor2 = drug2_data.get('HBondDonorCount', 0)
+    h_acceptor1 = drug1_data.get('HBondAcceptorCount', 0)
+    h_acceptor2 = drug2_data.get('HBondAcceptorCount', 0)
+    
+    if abs(h_donor1 - h_donor2) > 3 or abs(h_acceptor1 - h_acceptor2) > 5:
+        descriptions.append("Differing hydrogen bonding capacity may affect protein binding and metabolism")
     
-    # Based on hydrogen bonding
-    hbond_total1 = drug1_data.get('HBondDonorCount', 0) + drug1_data.get('HBondAcceptorCount', 0)
-    hbond_total2 = drug2_data.get('HBondDonorCount', 0) + drug2_data.get('HBondAcceptorCount', 0)
-    if abs(hbond_total1 - hbond_total2) > 8:
-        descriptions.append("Differing hydrogen bonding capacity may affect protein binding")
+    # Rotatable bonds analysis
+    rotatable1 = drug1_data.get('RotatableBondCount', 0)
+    rotatable2 = drug2_data.get('RotatableBondCount', 0)
+    
+    if abs(rotatable1 - rotatable2) > 8:
+        descriptions.append("Significant difference in molecular flexibility may influence receptor binding")
     
     # Default description if no specific features stand out
     if not descriptions:
-        descriptions.append("Potential pharmacokinetic interaction requiring clinical evaluation")
+        descriptions.append("Potential pharmacokinetic interaction based on molecular properties - clinical evaluation recommended")
     
-    return " ".join(descriptions)
+    return ". ".join(descriptions) + "."
 
 def check_drugbank_interaction(drug1_name, drug2_name):
-    """Placeholder function to check DrugBank interactions (you'd need API access)"""
-    # This is a mock function - you'd need actual DrugBank API access
+    """Mock function to simulate DrugBank interaction check"""
+    # Convert to lowercase for case-insensitive matching
     drug1_clean = drug1_name.lower().strip()
     drug2_clean = drug2_name.lower().strip()
     
-    # Mock known interactions for demonstration
+    # Expanded mock database of known interactions
     known_interactions = {
-        ('warfarin', 'aspirin'): 'Severe: Increased bleeding risk',
-        ('warfarin', 'ibuprofen'): 'Moderate: Increased bleeding risk',
-        ('simvastatin', 'clarithromycin'): 'Severe: Increased risk of myopathy',
-        ('digoxin', 'quinine'): 'Moderate: Increased digoxin levels',
+        ('warfarin', 'aspirin'): 'Severe: Increased risk of bleeding and hemorrhage',
+        ('warfarin', 'ibuprofen'): 'Moderate: Increased risk of gastrointestinal bleeding',
+        ('warfarin', 'simvastatin'): 'Moderate: Increased anticoagulant effect',
+        ('simvastatin', 'clarithromycin'): 'Severe: Increased risk of myopathy and rhabdomyolysis',
+        ('simvastatin', 'itraconazole'): 'Severe: Increased statin levels and toxicity risk',
+        ('digoxin', 'quinine'): 'Moderate: Increased digoxin levels, risk of toxicity',
+        ('digoxin', 'verapamil'): 'Moderate: Increased digoxin concentrations',
+        ('lisinopril', 'ibuprofen'): 'Moderate: Reduced antihypertensive effect',
+        ('metformin', 'contrast'): 'Severe: Risk of lactic acidosis with contrast media',
+        ('phenytoin', 'warfarin'): 'Moderate: Altered anticoagulant effect',
     }
     
-    # Check both orders
-    interaction = known_interactions.get((drug1_clean, drug2_clean)) or known_interactions.get((drug2_clean, drug1_clean))
+    # Check both orders of drug names
+    interaction = (known_interactions.get((drug1_clean, drug2_clean)) or 
+                   known_interactions.get((drug2_clean, drug1_clean)))
     
     if interaction:
         return interaction
     else:
-        return "No known interaction in mock database"
+        return "No known severe interaction in database (mock data)"
 
 def predict_ddi(drug1_name, drug2_name):
     """Main prediction function"""
     try:
+        if not drug1_name or not drug2_name:
+            return "Please enter both drug names", "", "", "", ""
+        
         # Fetch data for both drugs
         drug1_data, error1 = fetch_pubchem_data(drug1_name)
         drug2_data, error2 = fetch_pubchem_data(drug2_name)
         
-        if error1 or error2:
-            return f"Error: {error1 or error2}", "", "", "", ""
+        if error1:
+            return f"Error with {drug1_name}: {error1}", "", "", "", ""
+        if error2:
+            return f"Error with {drug2_name}: {error2}", "", "", "", ""
         
         # Generate interaction description
         interaction_description = generate_interaction_description(drug1_data, drug2_data)
@@ -119,35 +162,43 @@ def predict_ddi(drug1_name, drug2_name):
         
         # Prepare detailed output
         drug1_info = f"""
-        **{drug1_name} Properties:**
-        - SMILES: {drug1_data['CanonicalSMILES']}
-        - Molecular Weight: {drug1_data['MolecularWeight']:.2f}
-        - LogP: {drug1_data.get('XLogP', 'N/A')}
-        - TPSA: {drug1_data.get('TPSA', 'N/A')}
-        - H-Bond Donors: {drug1_data.get('HBondDonorCount', 'N/A')}
-        - H-Bond Acceptors: {drug1_data.get('HBondAcceptorCount', 'N/A')}
+        **{drug1_name}** (PubChem CID: {drug1_data.get('CID', 'N/A')})
+        - **IUPAC Name:** {drug1_data.get('IUPACName', 'N/A')}
+        - **SMILES:** {drug1_data.get('CanonicalSMILES', 'N/A')}
+        - **Molecular Weight:** {drug1_data.get('MolecularWeight', 'N/A')} g/mol
+        - **LogP (XLogP):** {drug1_data.get('XLogP', 'N/A')}
+        - **TPSA:** {drug1_data.get('TPSA', 'N/A')} Ų
+        - **H-Bond Donors:** {drug1_data.get('HBondDonorCount', 'N/A')}
+        - **H-Bond Acceptors:** {drug1_data.get('HBondAcceptorCount', 'N/A')}
+        - **Rotatable Bonds:** {drug1_data.get('RotatableBondCount', 'N/A')}
+        - **Complexity:** {drug1_data.get('Complexity', 'N/A')}
         """
         
         drug2_info = f"""
-        **{drug2_name} Properties:**
-        - SMILES: {drug2_data['CanonicalSMILES']}
-        - Molecular Weight: {drug2_data['MolecularWeight']:.2f}
-        - LogP: {drug2_data.get('XLogP', 'N/A')}
-        - TPSA: {drug2_data.get('TPSA', 'N/A')}
-        - H-Bond Donors: {drug2_data.get('HBondDonorCount', 'N/A')}
-        - H-Bond Acceptors: {drug2_data.get('HBondAcceptorCount', 'N/A')}
+        **{drug2_name}** (PubChem CID: {drug2_data.get('CID', 'N/A')})
+        - **IUPAC Name:** {drug2_data.get('IUPACName', 'N/A')}
+        - **SMILES:** {drug2_data.get('CanonicalSMILES', 'N/A')}
+        - **Molecular Weight:** {drug2_data.get('MolecularWeight', 'N/A')} g/mol
+        - **LogP (XLogP):** {drug2_data.get('XLogP', 'N/A')}
+        - **TPSA:** {drug2_data.get('TPSA', 'N/A')} Ų
+        - **H-Bond Donors:** {drug2_data.get('HBondDonorCount', 'N/A')}
+        - **H-Bond Acceptors:** {drug2_data.get('HBondAcceptorCount', 'N/A')}
+        - **Rotatable Bonds:** {drug2_data.get('RotatableBondCount', 'N/A')}
+        - **Complexity:** {drug2_data.get('Complexity', 'N/A')}
         """
         
         prediction_output = f"""
+        ## 🔍 AI Prediction Results
+        
         **Generated Interaction Description:**
-        {interaction_description}
+        *"{interaction_description}"*
         
-        **AI Prediction:**
-        - Severity: **{result['prediction']}**
-        - Confidence: {result['confidence']:.2%}
+        **Prediction:**
+        - **Severity:** **{result['prediction']}**
+        - **Confidence:** {result['confidence']:.1%}
         
-        **Probabilities:**
-        {', '.join([f'{k}: {v:.2%}' for k, v in result['probabilities'].items()])}
+        **Probability Distribution:**
+        {', '.join([f'{k}: {v:.1%}' for k, v in result['probabilities'].items()])}
         """
         
         return prediction_output, drug1_info, drug2_info, drugbank_result, interaction_description
@@ -158,49 +209,47 @@ def predict_ddi(drug1_name, drug2_name):
 # Create Gradio interface
 with gr.Blocks(title="Drug Interaction Severity Predictor", theme=gr.themes.Soft()) as demo:
     gr.Markdown("# 🧪 Drug Interaction Severity Predictor")
-    gr.Markdown("Predict potential drug-drug interaction severity using molecular properties and AI")
+    gr.Markdown("Predict potential drug-drug interaction severity using molecular properties from PubChem and AI")
     
     with gr.Row():
-        with gr.Column():
+        with gr.Column(scale=1):
+            gr.Markdown("## 💊 Input Drug Names")
             drug1 = gr.Textbox(label="First Drug Name", placeholder="e.g., Warfarin, Aspirin, Simvastatin...")
-            drug2 = gr.Textbox(label="Second Drug Name", placeholder="e.g., Ibuprofen, Clarithromycin...")
-            predict_btn = gr.Button("Predict Interaction", variant="primary")
+            drug2 = gr.Textbox(label="Second Drug Name", placeholder="e.g., Ibuprofen, Clarithromycin, Digoxin...")
+            predict_btn = gr.Button("🔬 Predict Interaction", variant="primary", size="lg")
+            
+            gr.Markdown("### 💡 Example Pairs")
+            gr.Examples(
+                examples=[
+                    ["Warfarin", "Aspirin"],
+                    ["Simvastatin", "Clarithromycin"],
+                    ["Digoxin", "Quinine"],
+                    ["Lisinopril", "Ibuprofen"]
+                ],
+                inputs=[drug1, drug2],
+                label="Try these examples:"
+            )
     
     with gr.Row():
-        with gr.Column():
+        with gr.Column(scale=2):
             gr.Markdown("## 📊 Prediction Results")
             prediction_output = gr.Markdown(label="AI Prediction")
             
+        with gr.Column(scale=1):
+            gr.Markdown("## 🏥 DrugBank Check")
+            drugbank_result = gr.Textbox(label="Known Interaction (Mock Data)", interactive=False)
+    
+    with gr.Row():
         with gr.Column():
-            gr.Markdown("## 🔍 Drug Properties")
-            with gr.Tab("Drug 1"):
+            gr.Markdown("## 🔍 Molecular Properties")
+            with gr.Tab("Drug 1 Properties"):
                 drug1_info = gr.Markdown()
-            with gr.Tab("Drug 2"):
+            with gr.Tab("Drug 2 Properties"):
                 drug2_info = gr.Markdown()
     
     with gr.Row():
-        with gr.Column():
-            gr.Markdown("## 🏥 DrugBank Comparison")
-            drugbank_result = gr.Textbox(label="Known Interaction (Mock Data)", interactive=False)
-        
-        with gr.Column():
-            gr.Markdown("## 📝 Generated Description")
-            interaction_description = gr.Textbox(label="AI-Generated Interaction Description", interactive=False)
-    
-    # Examples
-    gr.Markdown("### 💡 Example Drug Pairs")
-    gr.Examples(
-        examples=[
-            ["Warfarin", "Aspirin"],
-            ["Simvastatin", "Clarithromycin"],
-            ["Digoxin", "Quinine"],
-            ["Metformin", "Ibuprofen"]
-        ],
-        inputs=[drug1, drug2],
-        outputs=[prediction_output, drug1_info, drug2_info, drugbank_result, interaction_description],
-        fn=predict_ddi,
-        cache_examples=True
-    )
+        gr.Markdown("## 📝 Generated Description")
+        interaction_description = gr.Textbox(label="AI-Generated Interaction Description", interactive=False, lines=3)
     
     predict_btn.click(
         fn=predict_ddi,