merged pathways

epdb/models.py

@@ -2071,7 +2071,10 @@ class Pathway(EnviPathModel, AliasMixin, ScenarioMixin, AdditionalInformationMix
     def empty_due_to_threshold(self):
         return self.kv.get("empty_due_to_threshold", False)
 
-    def d3_json(self):
+    def merged_d3_json(self):
+        return self.d3_json(merged=True)
+
+    def d3_json(self, merged=False):
         # Ideally it would be something like this but
         # to reduce crossing in edges do a DFS
         # nodes = [n.d3_json() for n in self.nodes]
@@ -2079,15 +2082,68 @@ class Pathway(EnviPathModel, AliasMixin, ScenarioMixin, AdditionalInformationMix
         nodes = []
         processed = set()
 
+        structure_to_combined_nodes = dict()
+        node_to_combined_node = dict()
         queue = list()
-        for n in self.root_nodes:
-            queue.append(n)
 
-        # Add unconnected nodes
+        class CombinedNode(object):
-        for n in self.nodes.order_by("url"):
+
-            if len(n.out_edges.all()) == 0:
+            def __init__(self, representative: "Node"):
-                if n not in queue:
+                self.representative = representative
-                    queue.append(n)
+                self.nodes = set([self.representative])
+
+            def ids(self):
+                return [n.id for n in self.nodes]
+
+            def d3_json(self):
+                d3_json = self.representative.d3_json()
+                compound = self.representative.default_node_label.compound
+                d3_json["name"] = compound.get_name()
+                d3_json["smiles"] = compound.default_structure.smiles
+                d3_json["node_label_id"] = compound.default_structure.url
+                d3_json["image"] = f"{compound.default_structure.url}?image=svg",
+                d3_json["image_svg"] = IndigoUtils.mol_to_svg(
+                    compound.default_structure.molfile
+                    if compound.default_structure.molfile is not None
+                       and compound.default_structure.molfile.strip()
+                    else compound.default_structure.smiles,
+                    width=40,
+                    height=40,
+                ),
+
+                return d3_json
+
+            def __repr__(self):
+                return f"CombinedNode({self.nodes})"
+
+        if merged:
+            for n in self.nodes:
+                def_struct = n.default_node_label.compound.default_structure
+                if def_struct in structure_to_combined_nodes:
+                    structure_to_combined_nodes[def_struct].nodes.add(n)
+                else:
+                    cn = CombinedNode(n)
+                    structure_to_combined_nodes[def_struct] = cn
+
+            # for n in self.nodes.order_by("url"):
+            #     if len(n.out_edges.all()) == 0:
+            #         if n not in queue:
+            #             queue.append(n)
+
+            for cn in structure_to_combined_nodes.values():
+                for n in cn.nodes:
+                    node_to_combined_node[n] = cn
+
+            queue = list(structure_to_combined_nodes.values())
+        else:
+            for n in self.nodes:
+                queue.append(n)
+
+            # Add unconnected nodes
+            for n in self.nodes.order_by("url"):
+                if len(n.out_edges.all()) == 0:
+                    if n not in queue:
+                        queue.append(n)
 
         for i in queue:
             processed.add(i)
@@ -2097,25 +2153,56 @@ class Pathway(EnviPathModel, AliasMixin, ScenarioMixin, AdditionalInformationMix
             processed.add(current)
             nodes.append(current.d3_json())
 
-            for e in self.edges.filter(start_nodes=current).order_by("url").distinct():
+            if merged:
+                q = Q(start_nodes__in=current.ids())
+            else:
+                q = Q(start_nodes=current)
+
+            for e in self.edges.filter(q).order_by("url").distinct():
                 for prod in e.end_nodes.all().order_by("url"):
-                    if prod not in queue and prod not in processed:
+
-                        queue.append(prod)
+                    if merged:
+                        def_struct = n.default_node_label.compound.default_structure
+                        if def_struct in structure_to_combined_nodes:
+                            structure_to_combined_nodes[def_struct].nodes.add(n)
+                        else:
+                            cn = CombinedNode(n)
+                            structure_to_combined_nodes[def_struct] = cn
+                            if cn not in queue and cn not in processed:
+                                queue.append(cn)
+                    else:
+                        if prod not in queue and prod not in processed:
+                            queue.append(prod)
 
         # We shouldn't lose or make up nodes...
-        if len(nodes) != len(self.nodes):
+        if not merged:
-            logger.debug(
+            if len(nodes) != len(self.nodes):
-                f"{self.get_name()}: Num Nodes {len(nodes)} vs. DB Nodes {len(self.nodes)}"
+                logger.debug(
-            )
+                    f"{self.get_name()}: Num Nodes {len(nodes)} vs. DB Nodes {len(self.nodes)}"
+                )
 
         links = [e.d3_json() for e in self.edges]
 
         # D3 links Nodes based on indices in nodes array
+        combined_node_depth_to_idx = dict()
         node_url_to_idx = dict()
         for i, n in enumerate(nodes):
             n["id"] = i
             node_url_to_idx[n["url"]] = i
 
+        # Add potential missing mappings
+        if merged:
+            cn_idx_mapping = dict()
+            for n in nodes:
+                for cn in processed:
+                    for sub_n in cn.nodes:
+                        if n["url"] == sub_n.url:
+                            cn_idx_mapping[cn] = n["id"]
+
+            for n in processed:
+                for sub_n in n.nodes:
+                    node_url_to_idx[sub_n.url] = cn_idx_mapping[n]
+
         adjusted_links = []
         for link in links:
             # Check if we'll need pseudo nodes

epdb/views.py

@@ -2313,6 +2313,7 @@ def package_pathway(request, package_uuid, pathway_uuid):
 
         context["pathway"] = current_pathway
         context["current_object"] = current_pathway
+        context["merged"] = request.GET.get("merged", False) == "true"
 
         context["breadcrumbs"] = [
             {"Home": s.SERVER_URL},

templates/objects/pathway.html

@@ -544,8 +544,11 @@
       </div>
     {% endif %}
   </div>
+  {% if merged %}
+  {{ pathway.merged_d3_json|json_script:"pathway" }}
+  {% else %}
   {{ pathway.d3_json|json_script:"pathway" }}
-
+  {% endif %}
   <script>
     // Global switch for compound names view
     var compoundNamesViewEnabled = false;