Update app.py

app.py

@@ -226,15 +226,28 @@ def gradio_inference(dna_sequence, run_mapping):
 
     return summary, "\n".join(stats_lines), context_output, ""
 
+# --- CUSTOM CSS ---
+custom_css = """
+#scan_btn {
+    background-color: #0d6efd !important; /* Deep Blue */
+    color: white !important;
+    border: none !important;
+    transition: 0.3s ease;
+}
+#scan_btn:hover {
+    background-color: #dc3545 !important; /* Striking Red */
+}
+"""
+
 # --- THE UI LAYOUT ---
-with gr.Blocks(theme=gr.themes.Soft(), title="🧬 The Genomic Oracle 🧬") as demo:
+with gr.Blocks(theme=gr.themes.Soft(), title="🧬 The Genomic Oracle 🧬", css=custom_css) as demo:
     
-    # 1. The Custom HTML Title (Massive and Centered)
+    # 1. The Custom HTML Title
     gr.HTML(
         """
         <div style="text-align: center; padding-bottom: 10px;">
             <h1 style="font-size: 3.5rem; font-weight: bold; margin-bottom: 0.2rem;">🧬 The Genomic Oracle 🧬</h1>
-            <h3 style="margin-top: 0; font-weight: normal;"><b>University of Maryland Global Campus</b> | Bioinformatics Capstone</h3>
+            <h3 style="margin-top: 0; font-weight: normal;"><b>University of Maryland Global Campus (UMGC)</b> | Bioinformatics Capstone Project</h3>
         </div>
         <hr>
         """
@@ -246,23 +259,22 @@ with gr.Blocks(theme=gr.themes.Soft(), title="🧬 The Genomic Oracle 🧬") as
         Welcome to the official interface for **The Genomic Oracle**, a cascaded machine learning pipeline designed for high-precision DNA sequence classification. 
         
         ### The 4-Stage Cascading Architecture
-        1. **The Gatekeeper:** Logistic Regression model rapidly screens native k-mer vectors to identify protein-coding vs. non-coding potential.
-        2. **Structural Mapper:** LightGBM model classifies the sequence into 1 of 7 structural features (e.g., Exons, UTRs, Enhancers).
-        3. **Phenotype Prediction:** Sequences flagged as Coding are passed through a custom ALiBi BERT transformer model to predict specific traits.
-        4. **Regulatory Validation:** Sequences flagged as Non-Coding are routed to a DNABERT-2 spatial attention neural network.
+        1. **The Gatekeeper:** Logistic Regression Model rapidly screens raw k-mer vectors to identify protein-coding vs. non-coding potential.
+        2. **Structural Mapper:** LightGBM Model classifies the sequence into 1 of 7 structural features (e.g., Exons, UTRs, Enhancers).
+        3. **Phenotype Prediction:** Sequences flagged as Coding are passed through a custom ALiBi BERT transformer to predict specific traits.
+        4. **Regulatory Validation:** Sequences flagged as Non-Coding Promoters are routed to a DNABERT-2 spatial attention neural network.
+        
+        Created by: Kadir Galindo, Duncan Hall, Rebecca Mellinger & George Paccione
         ---
         """
     )
     
     with gr.Row():
-        # width of textbox: Both columns are now scale=1, giving you a perfect 50/50 split!
         with gr.Column(scale=1):
-            dna_input = gr.Textbox(label="Enter DNA Sequence", placeholder="e.g., ATGCGATCGATCGATCG...", lines=10)
+            dna_input = gr.Textbox(label="Enter DNA Sequence", placeholder="e.g., ATGCGATCGATCGATCG...", lines=12)
             run_mapping_cb = gr.Checkbox(value=False, label="Query NCBI BLAST for spatial mapping (Takes 1–3 mins)")
-            # 🚨 Changed variant="primary" to variant="stop" for a striking red button
-            submit_btn = gr.Button("🚀 Initialize Deep Scan", variant={"primary", "stop"})
+            submit_btn = gr.Button("🚀 Initialize Deep Scan", elem_id="scan_btn")
 
-        # width of textbox: Changed from scale=2 to scale=1
         with gr.Column(scale=1):
             output_summary = gr.Textbox(label=" Classification Summary", lines=4)
             stats_panel = gr.Textbox(label=" Internal Pipeline Statistics", lines=4)