Update app.py

app.py

@@ -2,76 +2,81 @@ import gradio as gr
 import torch
 import pickle
 import subprocess
+import pandas as pd
 from predictor import Predictor
 from tensorflow.keras.models import load_model
 from ml_simplified_tree import maximum_likelihood
 
-# --------- Load Models & Assets ---------
-# Boundary-aware model
+# --------- Load Models ---------
 boundary_model = Predictor("best_boundary_aware_model.pth")
-
-# Keras model
 keras_model = load_model("best_model.keras")
+
 with open("kmer_to_index.pkl", "rb") as f:
     kmer_to_index = pickle.load(f)
 
-# Utility function for keras model
+# --------- Utilities ---------
 def predict_with_keras(sequence):
     kmers = [sequence[i:i+6] for i in range(len(sequence)-5)]
     indices = [kmer_to_index.get(kmer, 0) for kmer in kmers]
     input_arr = torch.tensor([indices])
-    prediction = keras_model.predict(input_arr)
-    return str(prediction)
+    prediction = keras_model.predict(input_arr)[0]
+    return "".join(str(round(p, 3)) for p in prediction)
+
+def save_to_fasta(name, sequence, path):
+    with open(path, "w") as f:
+        f.write(f">{name}\n{sequence}\n")
 
-# MAFFT + IQTree pipeline
-def run_mafft_and_iqtree():
+def save_to_csv(sequence, path):
+    df = pd.DataFrame({"Sequence": [sequence]})
+    df.to_csv(path, index=False)
+
+def run_mafft_and_iqtree(fasta_file="f_gene_sequences_aligned.fasta"):
     try:
-        subprocess.run(["mafft", "--auto", "f_gene_sequences_aligned.fasta"], check=True)
+        subprocess.run(["mafft", "--auto", fasta_file], check=True)
         subprocess.run(["iqtree", "-s", "f_gene_sequences.phy.treefile", "-m", "GTR"], check=True)
         return "MAFFT and IQTree executed successfully."
     except Exception as e:
-        return f"Error: {str(e)}"
+        return f"Error running alignment/tree: {e}"
+
+def run_full_pipeline(dna_input):
+    # 1. Boundary-Aware Prediction
+    step1_out = boundary_model.predict(dna_input)
 
-# Maximum Likelihood Tree
-def run_maximum_likelihood():
+    # 2. Keras Prediction
+    step2_out = predict_with_keras(step1_out)
+
+    # 3. Save intermediate files
+    save_to_fasta("Predicted_Seq", step2_out, "f_gene_sequences_aligned.fasta")
+    save_to_csv(step2_out, "f gene clean dataset.csv")
+
+    # 4. Run MAFFT + IQTree
+    mafft_status = run_mafft_and_iqtree()
+
+    # 5. Run ML tree
     try:
-        result = maximum_likelihood("f gene clean dataset.csv")
-        return result
+        ml_output = maximum_likelihood("f gene clean dataset.csv")
     except Exception as e:
-        return f"Error: {str(e)}"
+        ml_output = f"ML Tree Error: {e}"
 
-# --------- Gradio Interface ---------
-def boundary_model_predict(sequence):
-    return boundary_model.predict(sequence)
+    return {
+        "Boundary Model Output": step1_out,
+        "Keras Model Output": step2_out,
+        "MAFFT + IQTree Status": mafft_status,
+        "Maximum Likelihood Tree Output": ml_output
+    }
 
-gr_interface = gr.TabbedInterface(
-    [
-        gr.Interface(
-            fn=boundary_model_predict,
-            inputs=gr.Textbox(label="DNA Sequence"),
-            outputs=gr.Textbox(label="Boundary Model Prediction"),
-            title="Boundary-Aware Model"
-        ),
-        gr.Interface(
-            fn=predict_with_keras,
-            inputs=gr.Textbox(label="DNA Sequence"),
-            outputs=gr.Textbox(label="Keras Model Prediction"),
-            title="Keras Model"
-        ),
-        gr.Interface(
-            fn=run_mafft_and_iqtree,
-            inputs=[],
-            outputs="text",
-            title="MAFFT + IQTree Runner"
-        ),
-        gr.Interface(
-            fn=run_maximum_likelihood,
-            inputs=[],
-            outputs="text",
-            title="Simplified ML Tree Generator"
-        )
+# --------- Gradio Interface ---------
+gr_interface = gr.Interface(
+    fn=run_full_pipeline,
+    inputs=gr.Textbox(label="Input DNA Sequence"),
+    outputs=[
+        gr.Textbox(label="Boundary Model Output"),
+        gr.Textbox(label="Keras Model Output"),
+        gr.Textbox(label="MAFFT + IQTree Status"),
+        gr.Textbox(label="ML Tree Output")
     ],
-    tab_names=["Boundary Model", "Keras Model", "MAFFT + IQTree", "ML Tree"]
+    title="Sequential Phylogenetic Inference Pipeline",
+    description="This pipeline runs sequentially: Boundary-Aware Model → Keras Model → Tree Building"
 )
 
 # --------- Launch ---------