added ratio as parameter

Author: ctrlaltaf

main.py

@@ -83,6 +83,7 @@ def main():
     namespace = ["molecular_function", "biological_process", "cellular_component"]
     sample_size = 1000
     repeats = 1
+    ratio = 100
     new_random_lists = True
     print_graphs = True
     no_inferred_edges = False
@@ -180,12 +181,17 @@ def main():
         # "HypergeometricDistributionV2": HypergeometricDistributionV2,
         "HypergeometricDistributionV3": HypergeometricDistributionV3,
         # "RandomWalk": RandomWalk,
-        "RandomWalkV2": RandomWalkV2,
+        # "RandomWalkV2": RandomWalkV2,
         # "RandomWalkV3": RandomWalkV3,
         # "RandomWalkV4": RandomWalkV4,   #need protein-only network
         # "RandomWalkV5": RandomWalkV5,     #need protein-goterm only network
     }
 
+    # for PW paper, we used OneHopGODegree, ProteinDegreeV3, HypergeometricDistributionV3, and RandomWalkV2
+    # with inferred and non inferred GO terms
+    # with complete network and ProGo only network
+    # using ratio of 1:1000
+
     run_workflow(
         algorithm_classes,
         go_protein_pairs,
@@ -199,6 +205,7 @@ def main():
         new_random_lists,
         short_name,
         print_graphs,
+        ratio
     )
 
     sys.exit()

tools/workflow.py

@@ -35,6 +35,7 @@ def run_workflow(
     new_random_lists,
     name,
     figure,
+    ratio
 ):
     """
     With a given set of algorithms, test the algorithms ability to prediction protein function on a given number of
@@ -88,6 +89,7 @@ def run_workflow(
                 dataset_directory_path,
                 i,
                 name,
+                ratio
             )
 
     for i in range(
@@ -712,7 +714,7 @@ def sample_data_neighbor_degree(
     return positive_dataset, negative_dataset
 
 def sample_data_neighbor_degree_ratio(
-    go_protein_pairs, sample_size, protein_list, G, input_directory_path, num, name
+    go_protein_pairs, sample_size, protein_list, G, input_directory_path, num, name, ratio
 ):
     """
     Given a sample size, generate positive nad negative datasets. positive dataset it generated by finding positive protein-goterm edges and for
@@ -735,7 +737,6 @@ def sample_data_neighbor_degree_ratio(
     positive_dataset = {"protein": [], "go": []}
     temp_positive_dataset = {"protein": [], "go": []}
     negative_dataset = {"protein": [], "go": []}
-    ratio = 100
 
     # sample the data
     for edge in sample(list(go_protein_pairs), sample_size):