Update model description in Gradio App

app.py

@@ -68,36 +68,40 @@ demo = gr.Interface(fn=classify,
     
     
 description_md = """
-### ℹ️ **About AllerTrans – Allergenicity Prediction Tool**
+### 📌 **About AllerTrans – An Allergenicity Prediction Tool for Protein Sequences**
 
 **🧬 Input Format – FASTA Sequences**  
 This tool accepts protein sequences in FASTA format
 
+**🧾 Output Explanation**
+
+AllerTrans classifies your input sequence into one of the following categories:
+
+🟢 Non-Allergen:
+The protein is unlikely to cause an allergic reaction and can be considered safe in terms of allergenicity.
+
+🔴 Potential Allergen:
+The protein has the potential to trigger an allergic response or exhibit cross-reactivity in certain individuals. While not all individuals may experience reactions, these proteins cannot be considered safe.
+
 **💡 Accepted Proteins**  
-- Natural and recombinant proteins  
-- Pharmaceutical and industrial proteins  
-- Synthetic sequences (tags or mutations allowed)
+- Natural and also recombinant proteins  
 
 🔎 **Note of Caution**:  
 While our model demonstrates promising performance—particularly with recombinant proteins, as evidenced by our additional evaluation with a recombinant protein dataset  
-from UniProt—**we advise caution when generalizing the results to all recombinant protein scenarios**.  
-The specificity of the model to various recombinant constructs and modifications has not been explored.
-
-**🧠 Prediction Process**  
-- Embeddings via ProtT5 + ESM-2  
-- Deep neural network for classification
+from UniProt—**we advise caution when generalizing the results to all constructs and modifications of recombinant protein**.  The generizability of the model to various recombinant scenarios has not been fully explored.
 
 **⚠️ Disclaimer**  
 Although AllerTrans provides highly accurate predictions, it is intended as a screening tool. For clinical or regulatory decisions, always confirm results with experimental validation.
 """
-
+    
 with gr.Blocks() as demo:
+    interface = gr.Interface(
+        fn=classify,
+        inputs=gr.Textbox(lines=3, placeholder="Enter protein sequence..."),
+        outputs=gr.Label(label="Prediction"),
+    )
+    interface.render()
     gr.Markdown(description_md)
-    with gr.Row():
-        input_box = gr.Textbox(lines=3, placeholder="Enter protein sequence...")
-        output_label = gr.Label(label="Prediction")
-    classify_btn = gr.Button("Run Prediction")
-    classify_btn.click(classify, inputs=input_box, outputs=output_label)
 
 if __name__ == "__main__":
     demo.launch()