add support for setting thresholds

app.py

@@ -14,11 +14,19 @@ if "gene_ids" in query_params.keys():
     input_gene_ids = query_params["gene_ids"]
 else:
     input_gene_ids = "TGME49_231630,TGME49_230210"
-
+    
 # use "\n" as the separator so it shows correctly in the text area
 input_gene_ids = input_gene_ids.replace(",", "\n")
-    
 
+if "coexp_score_threshold" in query_params.keys():
+    coexp_score_threshold = query_params["coexp_score_threshold"]        
+else:
+    coexp_score_threshold = "0.85"
+
+if "max_per_gene" in query_params.keys():
+    max_per_gene = query_params["max_per_gene"]        
+else:
+    max_per_gene = "25"
 
 st.markdown("""
 # ToxoCEN Network
@@ -54,7 +62,32 @@ with col1:
         value = f"{input_gene_ids}",
         help = "TGME49 Gene IDs e.g. TGME49_231630")
 
-coexp_score_threshold = 0.85
+
+with col2:
+    coexp_score_threshold = st.text_input(
+        label = "Co-expression threshold [0-1]",
+        value = f"{coexp_score_threshold}",
+        help = "Default: 0.85")
+
+    try:
+        coexp_score_threshold = float(coexp_score_threshold)
+    except:
+        st.error(f"Co-expression threshold should be a number between 0 and 1, instead it is '{coexp_score_threshold}'")
+    if coexp_score_threshold < 0 or 1 < coexp_score_threshold:
+        st.error(f"Co-expression threshold should be a number between 0 and 1, instead it is '{coexp_score_threshold}'")
+
+    max_per_gene = st.text_input(
+        label = "Max per gene",
+        value = f"{max_per_gene}",
+        help = "Default: 25")
+
+    try:
+        max_per_gene = int(max_per_gene)
+    except:
+        st.error(f"Max per gene should be a number greater than 0, instead it is '{max_per_gene}'")
+    if max_per_gene <= 0:
+        st.error(f"Max per gene should be a number greater than 0, instead it is '{max_per_gene}'")
+    
     
 ##################################
 # Parse and check the user input #
@@ -70,12 +103,14 @@ for input_gene_id in input_gene_ids.split("\n"):
 
 neighbors = []
 for seed_gene_id in seed_gene_ids:
-    neighbors.append(
-        top_coexp_hits[
+    hits = top_coexp_hits[
             (top_coexp_hits.gene_id_1 == seed_gene_id) & (top_coexp_hits.coexp_score > coexp_score_threshold)])
+    if len(hits.index) > max_per_gene:
+        hits = hits[0:max_per_gene]
+    neighbors.append(hits)
 
 neighbors = pd.concat(neighbors)
-
+    
 neighbor_gene_ids = list(set(neighbors.gene_id_2))
 gene_ids = seed_gene_ids + neighbor_gene_ids
 gene_types = ['seed'] * len(seed_gene_ids) + ['neighbor'] * len(neighbor_gene_ids)