Make adding the graph to docker simpler

docker/Dockerfile

@@ -29,7 +29,6 @@ RUN wget -O /sw/nodes.tsv.gz https://askem-mira.s3.amazonaws.com/dkg/$domain/bui
     sed -i 's/#dbms.security.auth_enabled/dbms.security.auth_enabled/' /etc/neo4j/neo4j.conf && \
     neo4j-admin import --delimiter='TAB' --skip-duplicate-nodes=true --skip-bad-relationships=true --nodes /sw/nodes.tsv.gz --relationships /sw/edges.tsv.gz
 
-COPY generate_obo_graphs.py /sw/generate_obo_graphs.py
 # Python packages
 RUN python -m pip install --upgrade pip && \
     python -m pip install git+https://github.com/gyorilab/mira.git@main#egg=mira[web,uvicorn,dkg-client,dkg-construct] && \
@@ -40,9 +39,10 @@ RUN python -m pip install --upgrade pip && \
     python -m pip install --no-dependencies "lxml>=4.6.4" && \
     python -m pip install --no-dependencies --ignore-requires-python sbmlmath
 
+RUN python -m mira.dkg.generate_obo_graphs
+
 # Copy the example json for reconstructing the ode semantics
 RUN wget -O /sw/sir_flux_span.json https://raw.githubusercontent.com/gyorilab/mira/main/tests/sir_flux_span.json
 
-RUN mkdir -p /graphs
 COPY startup.sh startup.sh
 ENTRYPOINT ["/bin/bash", "/sw/startup.sh"]

docker/README.md

@@ -40,13 +40,6 @@ docker run -p 8771:8771 -p 7687:7687 -e MIRA_NEO4J_URL=bolt://0.0.0.0:7687 mira:
 
 This exposes a REST API at `http://localhost:8771`. This also exposes Neo4j's bolt port at port 7687.
 
-
-Running the `build_run_docker.sh` script builds the docker image, 
-create directory `docker/mounted_graph_storage` to store the pickled obo 
-graphs, and start the container. When you first run the script and 
-start the container, it will take a few minutes to generate and store the 
-pickled graphs. 
-
 ## MIRA Metaregistry
 
 The MIRA metaregistry contains the prefixes and their associated metadata for all use cases.

mira/dkg/construct.py

@@ -460,10 +460,10 @@ def extract_ontology_subtree(curie: str, add_subtree: bool = False):
     resource_prefix = curie.split(":")[0]
     if resource_prefix == "ncbitaxon":
         type = "class"
-        cached_relabeled_obo_graph_path = (Path(__file__).resolve().parents[2]
-                                           / "docker" /
-                                           "mounted_graph_storage" /
-                                           "ncbitaxon_obo_graph.pkl")
+        version = get_version(resource_prefix)
+        cached_relabeled_obo_graph_path = prefix_directory_join(resource_prefix,
+                                                            name="relabeled_obo_graph.pkl",
+                                                            version=version)
 
         with open(cached_relabeled_obo_graph_path,'rb') as relabeled_graph_file:
             relabeled_graph = pickle.load(relabeled_graph_file)

mira/dkg/generate_obo_graphs.py

@@ -0,0 +1,34 @@
+from pyobo import get_version
+from pyobo.getters import _ensure_ontology_path
+from pyobo.utils.path import prefix_directory_join
+from obonet import read_obo
+import networkx
+import pickle
+
+
+def download_convert_ncbitaxon_obo_to_graph():
+    resource_prefix = "ncbitaxon"
+    version = get_version(resource_prefix)
+
+    # Checks to see if the pickled ncbitaxon obo graph exists in the container
+    cached_relabeled_obo_graph_path = prefix_directory_join(resource_prefix,
+                                                            name="relabeled_obo_graph.pkl",
+                                                            version=version)
+    if not cached_relabeled_obo_graph_path.exists():
+        _, obo_path = _ensure_ontology_path(
+            resource_prefix, force=False, version=version
+        )
+        obo_graph = read_obo(obo_path)
+
+        # Normalize node indices
+        relabeled_graph = networkx.relabel_nodes(
+            obo_graph, lambda node_index: node_index.lower()
+        )
+        with open(
+            cached_relabeled_obo_graph_path, "wb"
+        ) as relabeled_graph_file:
+            pickle.dump(relabeled_graph, relabeled_graph_file)
+
+
+if __name__ == "__main__":
+    download_convert_ncbitaxon_obo_to_graph()