data and package update

Dockerfile

@@ -12,18 +12,15 @@ RUN apt-get install -y supervisor nginx
 RUN apt-get install -y libgtk-3-dev
 RUN apt-get install wget
 
-COPY ./requirements.txt /usr/src/drugstone/requirements.txt
+RUN conda install -y conda python=3.8
+RUN conda install -c conda-forge -y graph-tool=2.45
 
-RUN conda install -y conda=4.3.16
-RUN conda install -c conda-forge -y graph-tool=2.32
+RUN pip install gunicorn
 
-RUN pip install pyvcf
+COPY ./requirements.txt /usr/src/drugstone/requirements.txt
 RUN pip install -r /usr/src/drugstone/requirements.txt
-RUN pip install gunicorn
 
 COPY ./supervisord.conf /etc/supervisor/conf.d/supervisord.conf
-#COPY scripts/docker-entrypoint.sh /usr/src/drugstone/docker-entrypoint.sh
-# COPY ./scripts/ /usr/src/drugstone/scripts/
 COPY ./python_nedrex/ /usr/src/drugstone/python_nedrex/
 RUN pip install /usr/src/drugstone/python_nedrex/
 

scripts/docker-entrypoint.sh

@@ -4,7 +4,7 @@ python3 manage.py makemigrations drugstone
 python3 manage.py migrate
 python3 manage.py createfixtures
 python3 manage.py cleanuptasks
-python3 manage.py populate_db --update -a
+#python3 manage.py populate_db --update -a
 python3 manage.py make_graphs
 
 /usr/bin/supervisord -c "/etc/supervisor/conf.d/supervisord.conf"

tasks/multi_steiner.py

@@ -127,39 +127,52 @@ def multi_steiner(task_hook: TaskHook):
         tree_edges.append((gtu.find_vertex(g, prop=g.vertex_properties[node_name_attribute], match=source_name)[0], gtu.find_vertex(g, prop=g.vertex_properties[node_name_attribute], match=target_name)[0]))
     cost_first_tree = sum([weights[g.edge(source, target)] for source, target in tree_edges])
     # returned_nodes = set(int(gtu.find_vertex(g, prop=g.vertex_properties['name'], match=first_tree.vertex_properties["name"][node])[0]) for node in range(first_tree.num_vertices()))
+    print(f"Before gtu: Costs={cost_first_tree}")
     returned_nodes = set(int(gtu.find_vertex(g, prop=g.vertex_properties[node_name_attribute], match=first_tree.vertex_properties[node_name_attribute][node])[0]) for node in range(first_tree.num_vertices()))
-
-    
+    print(f"After gtu: {returned_nodes}")
+    print(num_trees)
     if num_trees > 1:
+        print("num_trees > 1")
         is_bridge = find_bridges(g)
+        print("found bridges")
         edge_filter = g.new_edge_property("boolean", True)
+        print("filtered edges")
         found_new_tree = True
         while len(tree_edges) > 0:
+            print(f"Tree edges length: {len(tree_edges)}")
             if found_new_tree:
                 task_hook.set_progress(float(num_found_trees + 2) / (float(num_trees + 3)), "Computing Steiner tree {} of {}.".format(num_found_trees + 1, num_trees))
             found_new_tree = False
             tree_edge = tree_edges.pop()
+            print("1")
             g_edge = g.edge(tree_edge[0], tree_edge[1])
             if not is_bridge[g_edge]:
+                print("2")
                 edge_filter[g_edge] = False
                 g.set_edge_filter(edge_filter)
                 next_tree = steiner_tree(g, seeds, seed_map, weights, hub_penalty > 0)
+                print("3")
                 next_tree_edges = set()
                 for next_tree_edge in next_tree.edges():
                     # source_name = next_tree.vertex_properties["name"][next_tree.vertex_index[next_tree_edge.source()]]
                     # target_name = next_tree.vertex_properties["name"][next_tree.vertex_index[next_tree_edge.target()]]
                     # next_tree_edges.add((gtu.find_vertex(g, prop=g.vertex_properties['name'], match=source_name)[0], gtu.find_vertex(g, prop=g.vertex_properties['name'], match=target_name)[0]))
+                    print("4")
                     source_name = next_tree.vertex_properties[node_name_attribute][next_tree.vertex_index[next_tree_edge.source()]]
                     target_name = next_tree.vertex_properties[node_name_attribute][next_tree.vertex_index[next_tree_edge.target()]]
                     next_tree_edges.add((gtu.find_vertex(g, prop=g.vertex_properties[node_name_attribute], match=source_name)[0],gtu.find_vertex(g, prop=g.vertex_properties[node_name_attribute], match=target_name)[0]))
                 cost_next_tree = sum([weights[g.edge(source, target)] for source, target in next_tree_edges])
                 if cost_next_tree <= cost_first_tree * ((100.0 + tolerance) / 100.0):
+                    print("5")
                     found_new_tree = True
                     num_found_trees += 1
                     for node in range(next_tree.num_vertices()):
+                        print("GTU again")
                         # returned_nodes.add(int(gtu.find_vertex(g, prop=g.vertex_properties['name'], match=next_tree.vertex_properties["name"][node])[0]))
                         returned_nodes.add(int(gtu.find_vertex(g, prop=g.vertex_properties[node_name_attribute],match=next_tree.vertex_properties[node_name_attribute][node])[0]))
+                        print("GTU done")
                     removed_edges = []
+                    print("6")
                     for source, target in tree_edges:
                         if not ((source, target) in set(next_tree_edges)) or ((target, source) in set(next_tree_edges)):
                             removed_edges.append((source, target))

tasks/util/find_bridges.py

@@ -4,13 +4,16 @@ __time = 0
 
 
 def __dfs_find_bridges(g, node, visited, disc, low, parent, is_bridge):
-    
+    print("Dfs_find_bridges")
     visited[node] = True
     global __time
     disc[node] = __time
     low[node] = __time
     __time += 1
-    
+
+    print(f"Checking neighbors of {g.vertex(node)}")
+    print(f"Degree = {g.vertex(node).out_degree()}")
+
     for nb in g.get_all_neighbors(node):
         if not visited[nb]:
             parent[nb] = node
@@ -18,12 +21,12 @@ def __dfs_find_bridges(g, node, visited, disc, low, parent, is_bridge):
             low[node] = min(low[node], low[nb])
             if low[nb] > disc[node]:
                 is_bridge[g.edge(node, nb)] = True
-        elif int(nb) != parent[node]:
+        elif int(nb) != parent[node]: #TODO can in theory be removed because
             low[node] = min(low[node], disc[nb])
 
 def find_bridges(g):
     r"""Finds all bridges in a graph."""
-    
+    print("Finding bridges")
     global __time
     __time = 0
     sys.setrecursionlimit(g.num_vertices() + 1)

tasks/util/read_graph_tool_graph.py

@@ -79,13 +79,25 @@ def read_graph_tool_graph(file_path, seeds, max_deg, include_indirect_drugs=Fals
 
     # Check that all seed seeds have been matched and throw error, otherwise.
     # print(deleted_nodes)
-    print(seed_ids)
-    for protein, found in is_matched.items():
-        if not found:
-            raise ValueError("Invalid seed protein {}. No node named {} in {}.".format(protein, protein, file_path))
+    # print(seed_ids)
+    # for protein, found in is_matched.items():
+    #     if not found:
+    #         raise ValueError("Invalid seed protein {}. No node named {} in {}.".format(protein, protein, file_path))
 
     # Delete edges that should be ignored or are not contained in the selected dataset.
     deleted_edges = []
+
+    for edge in g.edges():
+        if edge.source == edge.target:
+            deleted_edges.append(edge)
+
+    g.set_fast_edge_removal(fast=True)
+    for edge in deleted_edges:
+        g.remove_edge(edge)
+    g.set_fast_edge_removal(fast=False)
+
+    deleted_edges = []
+
     if (drug_ids and not include_indirect_drugs):  # If only_direct_drugs should be included, remove any drug-protein edges that the drug is not a direct neighbor of any seeds
         direct_drugs = set()
         for edge in g.edges():