Update app.py (#10)

- Update app.py (43eb70f73fd8688348085853d9672fb0ec306e74)


Co-authored-by: Yanis Adel <[email protected]>

app.py

@@ -22,61 +22,55 @@ class Log:
 
 # --- Main Function ---
 @spaces.GPU
-def run_chatnt(fasta_text, custom_question):
+def run_chatnt(dna_text, fasta_file, custom_question):
     with open(log_file, "a") as log:
         log.write("Request started\n\n")
 
-    if not custom_question or custom_question.strip() == "":
-        return "Please provide a question."
-
-    # Read DNA sequences from pasted text
-    dna_sequences = []
-    if fasta_text:
-        lines = fasta_text.splitlines()
+    # Read DNA sequence from text field or file
+    dna_sequence = ""
+    if dna_text and dna_text.strip():
+        dna_sequence = dna_text.strip().replace("\n", "")
+    elif fasta_file is not None:
+        file_content = fasta_file.read().decode("utf-8")
+        lines = file_content.splitlines()
         sequence = ""
         for line in lines:
             line = line.strip()
-            if not line:
+            if not line or line.startswith(">"):
                 continue
-            if line.startswith(">"):
-                if sequence:
-                    dna_sequences.append(sequence)
-                    sequence = ""
-            else:
-                sequence += line
-        if sequence:
-            dna_sequences.append(sequence)
-
-    if not dna_sequences:
-        return "No DNA sequences found in the input."
+            sequence += line
+        dna_sequence = sequence
+
+    dna_sequences = []
+    if dna_sequence:
+        dna_sequences.append(dna_sequence)
 
     with open(log_file, "a") as log:
-        for i, seq in enumerate(dna_sequences):
-            log.write(f"DNA sequence {i+1} : {seq}\n")
+        log.write(f"DNA sequences found: {dna_sequences}\n")
+
+    # Check DNA sequences count
+    if len(dna_sequences) > 1:
+        return "You must use only one DNA sequence."
+
+    if not custom_question or custom_question.strip() == "":
+        return "Please provide a question."
 
     # Build prompt
-    num_sequences = len(dna_sequences)
     num_placeholders = custom_question.count("<DNA>")
 
-    if num_sequences == 1:
+    if len(dna_sequences) == 0:
+        english_sequence = custom_question
+    else:
         if num_placeholders == 0:
-            english_sequence = custom_question + " <DNA>"
+            return "Your question must include the <DNA> token at the position where the DNA sequence should be inserted."
         elif num_placeholders == 1:
             english_sequence = custom_question
         else:
-            return "Too many <DNA> placeholders for a single DNA sequence."
-    elif num_sequences > 1:
-        if num_placeholders != num_sequences:
-            return f"You provided {num_sequences} DNA sequences but only {num_placeholders} <DNA> placeholders. Please specify one <DNA> for each sequence."
-        english_sequence = custom_question
-    else:
-        return "No DNA sequences detected."
-
-    with open(log_file, "a") as log:
-        log.write(f"Initial user question : {custom_question}\n")
-        log.write(f"Full english prompt : {english_sequence}\n")
+            return "You can only provide one DNA sequence, so you must use exactly one <DNA> placeholder."
 
     with open(log_file, "a") as log:
+        log.write(f"Initial user question: {custom_question}\n")
+        log.write(f"Full english prompt: {english_sequence}\n")
         log.write("Calling model\n")
 
     output = pipe(
@@ -86,8 +80,11 @@ def run_chatnt(fasta_text, custom_question):
         }
     )
 
+    if len(dna_sequences) == 0:
+        return f"{output}\n\nNote: Careful, you did not provide any DNA sequence."
+
     with open(log_file, "a") as log:
-        log.write(f"Output : {output}\n")
+        log.write(f"Output: {output}\n")
 
     return output
 
@@ -103,14 +100,20 @@ with gr.Blocks(css=css) as demo:
 
     with gr.Row():
         with gr.Column(scale=1):
-            fasta_text = gr.Textbox(
-                label="Paste your DNA sequences in FASTA format",
-                placeholder=">seq1\nATGC...\n>seq2\nCGTA...",
-                lines=8
+            dna_text = gr.Textbox(
+                label="Paste your DNA sequence",
+                placeholder="ATGCATGCATGC...",
+                lines=4
             )
+
+            fasta_file = gr.File(
+                label="Or upload your FASTA file",
+                file_types=[".fasta", ".fa", ".txt"]
+            )
+
             custom_question = gr.Textbox(
                 label="English Question",
-                placeholder="e.g., Does this sequence contain a donor splice site? <DNA>"
+                placeholder="e.g., Does this sequence <DNA> contain a donor splice site?"
             )
 
             submit_btn = gr.Button("Run Query", variant="primary")
@@ -120,14 +123,15 @@ with gr.Blocks(css=css) as demo:
 
     submit_btn.click(
         run_chatnt,
-        inputs=[fasta_text, custom_question],
+        inputs=[dna_text, fasta_file, custom_question],
         outputs=output,
     )
 
     gr.Markdown("""
-**Note:** Your question **must** include the `<DNA>` token if needed for multiple sequences.  
-Example if your FASTA text contains two sequences :  
-"Does the sequence <DNA> contain a donor splice site? And the sequence <DNA> ?"
+**Note:**  
+✅ You must use **exactly one DNA sequence** (either paste it or upload a file).  
+✅ Your question must include the `<DNA>` token **exactly once** at the position where the DNA will be inserted.  
+Example: *"Does this sequence <DNA> contain a donor splice site?"*
     """)
 
     with gr.Accordion("Logs", open=True):