Update ASCARIS.py

ASCARIS.py

@@ -4,28 +4,27 @@ from os import path
 import sys
 import streamlit.components.v1 as components
 sys.path.append('code/')
+#sys.path.append('ASCARIS/code/') 
 import pdb_featureVector
 import alphafold_featureVector
 import argparse
 from st_aggrid import AgGrid, GridOptionsBuilder, JsCode,GridUpdateMode
 import base64
-from huggingface_hub import hf_hub_download
-import streamlit as st
-import gzip
-
 showWarningOnDirectExecution = False
 
 def convert_df(df):
-   return df.to_csv(index=False, sep= '\t').encode('utf-8')
-
+   return df.to_csv(index=False).encode('utf-8')
 
+    
+# Check if 'key' already exists in session_state
+# If not, then initialize it
 if 'visibility' not in st.session_state:
     st.session_state['visibility'] = 'visible'
     st.session_state.disabled =  False
 
-original_title = '<p style="font-family:Trebuchet MS; color:#000000; font-size: 25px; font-weight:bold; text-align:center">ASCARIS</p>'
+original_title = '<p style="font-family:Trebuchet MS; color:#FD7456; font-size: 25px; font-weight:bold; text-align:center">ASCARIS</p>'
 st.markdown(original_title, unsafe_allow_html=True)
-original_title = '<p style="font-family:Trebuchet MS; color:#000000; font-size: 25px; font-weight:bold; text-align:center">(Annotation and StruCture-bAsed RepresentatIon of Single amino acid variations)</p>'
+original_title = '<p style="font-family:Trebuchet MS; color:#FD7456; font-size: 25px; font-weight:bold; text-align:center">(Annotation and StruCture-bAsed RepresentatIon of Single amino acid variations)</p>'
 st.markdown(original_title, unsafe_allow_html=True)
  
 st.write('')
@@ -35,49 +34,46 @@ st.write('')
 
 with st.form('mform', clear_on_submit=False):
     source = st.selectbox('Select the protein structure resource (1: PDB-SwissModel-Modbase, 2: AlphaFold)',[1,2])
-    impute = st.selectbox('Missing value imputation (mostly for the cases where the corresponding annotation does not exist in the protein)',[True, False])
-    input_data = st.text_input('Enter SAV data points (format: "UniProt/Swiss-Prot human protein accession" – "wild type a.a." – "position on the sequence" – "mutated a.a.").   Example: P04217-E-20-A or O43556-I-40-A,P57737-W-372-A')
-            
-
+    impute = st.selectbox('Imputation',[True, False])
+    
+    input_data = st.text_input('Enter SAV data points (Example: Q9BTP7-S-126-F, P04217-A-493-S, Q00889-G-2-L)')
 
     parser = argparse.ArgumentParser(description='ASCARIS')
-    
-    
+
     input_set = input_data
+    mode = source
     impute = impute
     submitted = st.form_submit_button(label="Submit", help=None, on_click=None, args=None, kwargs=None, type="secondary", disabled=False, use_container_width=False)
     print('*****************************************')
     print('Feature vector generation is in progress. \nPlease check log file for updates..')
     print('*****************************************')
-    mode = int(source)
+    mode = int(mode)
+    
     
 selected_df = pd.DataFrame()
-st.write('The online tool may be slow, especially while processing multiple SAVs. To address this, please consider using the programmatic version at https://github.com/HUBioDataLab/ASCARIS/')
+st.write('The online tool may be slow, especially while processing multiple SAVs and with multiple PDB matches, please consider using the local programmatic version at https://github.com/HUBioDataLab/ASCARIS/')
 if submitted:
     with st.spinner('In progress...This may take a while...'):
-        try:
-            if mode == 1:
-                selected_df = pdb_featureVector.pdb(input_set, mode, impute)   
-                
-            elif mode == 2:
-                selected_df = alphafold_featureVector.alphafold(input_set, mode, impute)
-            else:
-                selected_df =  pd.DataFrame()
-
-        except:
-            selected_df = pd.DataFrame()
-            pass
+        # try:
+        if mode == 1:
+            selected_df = pdb_featureVector.pdb(input_set, mode, impute)  
+        elif mode == 2:
+            selected_df = alphafold_featureVector.alphafold(input_set, mode, impute)
+        else:
+            selected_df =  pd.DataFrame()
 
     if selected_df is None:
-        st.success('Feature vector failed. Check the log file.')
+        st.success('No feature vector is created. Check the log file.')
 
     else:
         if len(selected_df) != 0 :
             st.write(selected_df)
-            st.success('Feature vector is successfully created.')
+            st.success('Feature vector successfully created.')
             csv = convert_df(selected_df)
+            
     
             st.download_button("Press to Download the Feature Vector", csv,f"ASCARIS_SAV_rep_{input_set}.csv","text/csv",key='download-csv')
 
         else:
-            st.success('Feature vector failed. Check the log file.')
+            st.success('Feature vector failed. Check log file.')
+