Add flanked CRISPR example for better prediction visualization

- Added FLANKED_CRISPR_EXAMPLE with ~500bp upstream + CRISPR array + ~500bp downstream
- Set flanked example as default in Predict & Visualize tab
- Reorganized example buttons: "CRISPR Array Only", "Flanked CRISPR (recommended)", "Non-CRISPR"
- Visualization now shows characteristic low-high-low curve pattern

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

app.py

@@ -49,6 +49,10 @@ CRISPR_EXAMPLE = """TCCCCATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTCTGTTTACTTCCCTCTA
 
 NON_CRISPR_EXAMPLE = """TTCGTTCATTTTTCTGGTTTGACCAATAGCATTTAAAGCCGCCCCACATAAATCATTTGGCCCGGAAACCTTTTGGTAAGATAAAATAGTGCCATCACTAGAAAGCTCAATTCTAATATCACAGACCTTACCAAAAAAACGACTATCACCACGAAAATATGGCTTCACAGATTTATAGATAATACCTGCATACCTCGAACCCGCCTTACCGCCATCGCCTGCGCCCGCTGAAGCACCTGAACCCTGGCTGCCTTGTTTATTTATGTTACTACCTTGCAAAGCACTGCCGCCGCCCACATCACCGCCATTAAAAAAATCATCTAATGCATTTTGATCAGCTCGACGTTGCGCATCCGCTTTCGCTTTGGCTTCCGCATCTGCCTTAGCCTTGGCGTCAGCTTTCGCTTTAGCATCGGCCTTAGCTTTTGCTTCTGCGTCAGCTTTCGCTTTAGCATCCGCTTTGGCTTTTGCATCAGCTTCCGCCTTCGCTTTAGCCTGCGCCTCTGCTTTTGCCTTAGCTTCCGCCTCTGCTTTGGCTTTTGCCTCTTGTTTTGCTTTCTCGTCTGCCTGCACTTTGGCTTTTGCTTCTTCTTCCGCTTGTTTTGCAGCATCTGCTAAACGTTTTGCCTCAGCTTCAGCTTTGAGCTTAGCGGCTTCAGCAGCCTGTTTGGCTTTCGCCTCTTCTGCCTGTTTTTGCTTTTCCAAGGCTTCTAAGCGAGCTTTTTCTTCCTGTTGCTTTTTCTGTTCAGCCAAGAACCTTTGTCTTTCCAGCTCTTTTTGCTGTTCCAGTTCTTTCTGACGGGCAATTTCCTGTTGACGTTGTTGCTCTTTTAACACTTCCCGTTGTTTTTCTTCTTCCCGTTTCTGCTCTTCACGTTTTTGGTCTTCAAGGGCTTGTTTCTTTTGTCTGTCCGCCTGACCTTTTTTCTGTTGTTGAATTCGCCCCCATTCCTGCGCTGCCGAGCCCGTATCAACCATCACGGCGCCGATAACTTCACCG"""
 
+# Flanked CRISPR example: upstream (500bp) + CRISPR array (10 repeats) + downstream (500bp)
+# This shows nice visualization with low score on flanks and high score in the middle
+FLANKED_CRISPR_EXAMPLE = """ATGCGATCGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATTCCCCATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTCTGTTTACTTCCCTCTATATCTTTTTTTGTTCGGTCATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTAAAATCACACTCACAGCCAATACAAGCGGGGGGGGAAATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTTGCAGTAGGGCAGACTGGCAGTTTTCGGGTAATGATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACATTCATACGAATAATCATTTCCGAAAGACTCCTTTTATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACAGGTCATGAGCATTCAAAACGTTCTCCCCGTTCAATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTAGCCTGGACCAAATAATGTACGAACCTCTCCATCTATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACATGAATTATATAACAGGGATTAAAATTTTTCTTATTATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTAAATTTGAGCAAATACTAAAAAAATGAGACAAAAAGATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTCCGGCAATGAATTGATAGGACTTAAAATAATTGTATTCGAGAGCAAGATCCACTAAAACAAGGATTGAAACTATCACGTTGAACGATCGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGATCGATCGATCGATCGATCGTAGCTAGCTAGCTAGCTAGCTGATCGATCGATCGTAGCTAGCTAGCTGAT"""
+
 # Longer examples for State-Dynamic Plot (upstream + CRISPR array + downstream)
 # Structure: ~600bp upstream | CRISPR array (25 repeats + 24 spacers) | ~600bp downstream
 # Total: ~3000 bp - ideal for seeing alternating patterns in State-Dynamic Plot
@@ -724,7 +728,7 @@ Detect CRISPR arrays in DNA sequences using a BERT-based deep learning model (43
                     label="DNA Sequence (min 1000 bp)",
                     placeholder="Paste DNA sequence or FASTA...",
                     lines=8,
-                    value=CRISPR_EXAMPLE
+                    value=FLANKED_CRISPR_EXAMPLE
                 )
                 with gr.Row():
                     stride_input = gr.Slider(
@@ -737,10 +741,14 @@ Detect CRISPR arrays in DNA sequences using a BERT-based deep learning model (43
                     )
                 with gr.Row():
                     predict_btn = gr.Button("Analyze Sequence", variant="primary")
-                    gr.Button("Load CRISPR Example").click(
+                with gr.Row():
+                    gr.Button("CRISPR Array Only").click(
                         lambda: CRISPR_EXAMPLE, outputs=seq_input
                     )
-                    gr.Button("Load Non-CRISPR Example").click(
+                    gr.Button("Flanked CRISPR (recommended)").click(
+                        lambda: FLANKED_CRISPR_EXAMPLE, outputs=seq_input
+                    )
+                    gr.Button("Non-CRISPR").click(
                         lambda: NON_CRISPR_EXAMPLE, outputs=seq_input
                     )
                 result_summary = gr.Markdown()