Update app.py

app.py

@@ -78,18 +78,13 @@ hits, fid_prom, pid_prom = load_hits()
 pid2name, name2pid, any2pid = pid2name_mapper()
 fid2smi = load_fid2smi()
 
+st.sidebar.title("Ligand Discovery 5: On-the-Fly Model")
+st.sidebar.write("this app builds a quick ML model for your proteins of interest")
 
-st.title("Ligand Discovery 5: On-the-Fly ML Models")
-st.write("input your proteins of interest and we'll build a quick ML model")
+st.sidebar.subheader(":mag: UniProt Accession IDs or Gene Name")
 
-cols = st.columns([1, 1, 2.4])
-col = cols[0]
-col.subheader(":mag: input your proteins")
+text = st.sidebar.text_area("For example, you can query VDAC2", placeholder = "VDAC2", help="Write one protein per line. UniProt AC format is preferred. Only proteins available in the Ligand Discovery interactome will be considered")
 
-text = col.text_area(
-    "input proteins in UniProt AC format or Gene Name. For example, you can query VDAC2",
-    help="Write one protein per line. UniProt AC format is preferred. Only proteins available in the Ligand Discovery interactome will be considered",
-)
 input_tokens = text.split()
 input_pids = []
 for it in input_tokens:
@@ -105,7 +100,7 @@ tfidf = True
 if input_data.shape[0] == 0:
     has_input = False
     if len(input_tokens) > 0:
-        col.warning(
+        st.sidebar.warning(
             "None of your input proteins was found in the Ligand Discovery interactome.".format(
                 len(input_pids), len(input_tokens)
             )
@@ -117,13 +112,13 @@ if has_input:
     print("Instantiating on the fly model")
     model = OnTheFlyModel()
     is_fitted = False
-    col.info(
+    st.sidebar.info(
         "{0} out of {1} input proteins were found in the Ligand Discovery interactome, corresponding to all statistically significant fragment-protein pairs.".format(
             len(input_pids), len(input_tokens)
         )
     )
 
-    col.dataframe(input_data, hide_index=True)
+    st.sidebar.dataframe(input_data, hide_index=True)
 
     uniprot_inputs = list(input_data["UniprotAC"])
     if len(uniprot_inputs) == 1:
@@ -150,7 +145,8 @@ if has_input:
             joblib.dump((graph_key, clusters_of_proteins), clusters_cache_file)
             old_clusters_cache_file = clusters_cache_file
 
-    col = cols[1]
+    cols = st.columns([0.3, 0.7])
+    col = cols[0]
 
     col.subheader(":robot_face: Quick modeling")
 
@@ -284,7 +280,7 @@ if has_input:
         is_ready = True
 
     if is_ready:
-        col = cols[2]
+        col = cols[1]
 
         col.subheader(":crystal_ball: Make predictions")
 
@@ -379,4 +375,4 @@ if has_input:
                 file_name="prediction_output.csv",
                 mime="text/csv",
             )
-            del model
+            del model