using plotly

app.py

@@ -4,6 +4,7 @@ import cas9on
 import cas9off
 import pandas as pd
 import streamlit as st
+import plotly.express as px
 import numpy as np
 from pathlib import Path
 from igv_component import igv_component
@@ -93,37 +94,47 @@ if selected_model == 'Cas9':
     )
 
     if target_selection == 'on-target':
+        # Gene symbol entry
         gene_symbol = st.text_input('Enter a Gene Symbol:', key='gene_symbol')
 
-        # Add genome reference selection
-        genome_reference = st.selectbox('Select a Genome Reference:', ['hg19', 'hg38'], key='genome_reference')
-
+        # Prediction button
         predict_button = st.button('Predict on-target')
 
+        # Process predictions
         if predict_button and gene_symbol:
             predictions = cas9on.process_gene(gene_symbol, cas9on_path)
-            st.session_state['on_target_results'] = predictions[:10]
-            st.session_state['full_on_target_results'] = predictions
+            # Sort predictions by the 'Prediction' score in descending order and take the top 10
+            sorted_predictions = sorted(predictions, key=lambda x: x[-1], reverse=True)[:10]
+            st.session_state['on_target_results'] = sorted_predictions
 
+        # On-target results display
         if 'on_target_results' in st.session_state and st.session_state['on_target_results']:
-            # Convert non-serializable types (like numpy float32) to Python native types
-            crispr_results_serializable = [
-                [item if not isinstance(item, np.floating) else float(item) for item in row]
-                for row in st.session_state['on_target_results']
-            ]
-
-            # Pass the serializable results to the IGV component
-            igv_component(crispr_results=crispr_results_serializable,
-                          genome_reference=genome_reference,
-                          gene_symbol=gene_symbol,
-                          key="igv_viewer")
-
+            # Convert the results to a pandas DataFrame for better display
+            df = pd.DataFrame(st.session_state['on_target_results'],
+                              columns=["Gene ID", "Start Pos", "End Pos", "Strand", "gRNA", "Prediction"])
+            st.write('Top on-target predictions:')
+            st.dataframe(df)
+
+            # Plotting with Plotly
+            fig = px.bar(df, x='gRNA', y='Prediction', color='Prediction',
+                         labels={'gRNA': 'Guide RNA', 'Prediction': 'Prediction Score'},
+                         hover_data=['Start Pos', 'End Pos', 'Strand'])
+            fig.update_layout(title='Top 10 CRISPR Targets by Prediction Score',
+                              xaxis_title='Guide RNA',
+                              yaxis_title='Prediction Score')
+            st.plotly_chart(fig)
+
+            # Download button for full results
             if 'full_on_target_results' in st.session_state:
                 full_df = pd.DataFrame(st.session_state['full_on_target_results'],
                                        columns=["Gene ID", "Start Pos", "End Pos", "Strand", "gRNA", "Prediction"])
                 full_predictions_csv = full_df.to_csv(index=False).encode('utf-8')
-                st.download_button(label='Download on-target predictions', data=full_predictions_csv,
-                                   file_name='on_target_results.csv', mime='text/csv')
+                st.download_button(
+                    label='Download all on-target predictions',
+                    data=full_predictions_csv,
+                    file_name='on_target_results.csv',
+                    mime='text/csv'
+                )
 
     elif target_selection == 'off-target':
         ENTRY_METHODS = dict(

requirements.txt

@@ -3,3 +3,4 @@ biopython==1.80
 pandas==1.5.2
 tensorflow==2.11.0
 tensorflow-probability==0.19.0
+plotly==5.18.0