bugfix for cache, more information in nodes

Author: ialbert

Makefile

@@ -5,15 +5,31 @@
 #
 
 # Default OBO file.
-URL = http://current.geneontology.org/ontology/go-basic.obo
-OBO = tmp/go-basic.obo
+OBO_URL = http://current.geneontology.org/ontology/go-basic.obo
+OBO_FILE = tmp/go-basic.obo
+
+# The gene association file.
+GAF_URL = https://current.geneontology.org/annotations/goa_human.gaf.gz
+GAF_FILE = tmp/goa_human.gaf.gz
 
 # Usage information.
 usage:
 	@echo "#"
 	@echo "# Use the source, Luke!"
 	@echo "#"
 
+# Download the gene ontology file
+${OBO_FILE}:
+	mkdir -p $(dir ${OBO_FILE})
+	curl -L ${OBO_URL} > ${OBO_FILE}
+
+${GAF_FILE}:
+	mkdir -p $(dir ${GAF_FILE})
+	curl -L ${GAF_URL} > ${GAF_FILE}
+
+index: ${OBO_FILE} ${GAF_FILE}
+	genescape build --obo ${OBO_FILE} --gaf ${GAF_FILE} --index ${GAF_FILE}.index.gz
+
 # Performs the testing.
 web:
 	python src/genescape/server.py
@@ -26,10 +42,6 @@ test:
 testall: test
 	(cd test && make test)
 
-# A full test with file generation.
-testall: test
-	(cd test && make test)
-
 # Runs a linter.
 lint:
 	hatch run lint:style

src/genescape/__about__.py

@@ -1 +1 @@
-__version__ = "0.8.0"
+__version__ = "0.8.1"

src/genescape/annot.py

@@ -33,7 +33,8 @@ def build():
         return idx
 
     # Cache the index
-    idx = resources.cache("index", func=build)
+    key = f"file_{index}"
+    idx = resources.cache(key, func=build)
 
     # Extract the gene names from the data
     names = map(lambda x: x[utils.GID], data[utils.DATA_FIELD])
@@ -109,7 +110,9 @@ def go2ns(goid):
     res = []
     data_fields = [utils.GID, "root", "count", "function", utils.SOURCE, "count", "size", utils.LABEL]
 
+
     for goid, cnt, func in counts:
+
         label = f"({cnt}/{n_size})"
         funcs = func2name.get(goid, [])
         name_space = go2ns(goid)
@@ -152,17 +155,21 @@ def ann2csv(ann):
 
 
 if __name__ == "__main__":
+
+    # Get default config
+    cnf = resources.get_config()
+
     # Initialize the resources
-    res = resources.init()
+    res = resources.init(cnf)
 
     inp = res.TEST_GENES
-    index = res.INDEX
+    ind = res.INDEX
 
     # Read the genelist
     stream = utils.get_stream(inp=inp)
 
     data = utils.parse_terms(iterable=stream)
 
-    out = run(data=data, index=index, csvout=True)
+    out = run(data=data, index=ind, csvout=True)
 
     print(out)

src/genescape/build.py

@@ -89,15 +89,33 @@ def make_index(obo, gaf, index, with_synonyms=False):
             sym2go.setdefault(sym, []).append(go_id)
             go2sym.setdefault(go_id, []).append(sym)
 
+    # Remove duplicates
+    for key in sym2go:
+        sym2go[key] = list(set(sym2go[key]))
+
+    for key in go2sym:
+        go2sym[key] = list(set(go2sym[key]))
+
     utils.info(f"reading: {obo}")
     stream = gzip.open(obo, mode="rt", encoding="utf-8") if obo.name.endswith(".gz") else open(obo)
     terms = parse_obo(stream)
     terms = filter(lambda x: not x.get("is_obsolete"), terms)
     terms = list(terms)
 
+    # The total number of symbols
+    sym_count = len(sym2go) or 1
+
+    # Number of genes annotated with this GO term.
+    def gene_count(goid):
+        return len(go2sym.get(goid, []))
+
     # Create a dictionary of GO terms
     obo_dict = {}
     for term in terms:
+        goid = term["id"]
+        gcnt = gene_count(goid)
+        term[utils.GO2GENE_COUNT] = gcnt
+        term[utils.GO2GENE_PERC] = (gcnt / sym_count) * 100
         obo_dict[term["id"]] = term
 
     # Database metadata

src/genescape/data/genescape.hs.json.gz

@@ -18,6 +18,10 @@
 def cache(key, func):
     global CACHE
 
+    #key = str(key)
+
+    utils.info(f"cache key: {key}")
+
     if key not in CACHE:
         CACHE[key] = func()
 
@@ -65,7 +69,7 @@ def get(name):
 
     def get_index(self, code):
         for value in self.config.get("index", []):
-            if value["code"] == code:
+            if value.get("code", "") == code:
                 return value["path"]
         utils.error("index code not found: {code}")
         return self.INDEX

src/genescape/data/genescape.mm.json.gz

@@ -117,6 +117,8 @@ def draw():
         count = safe_int(count, default=1)
         db = get_param(request, name='db')
 
+        #resources.clear_cache()
+
         # Choose a different index
         index = res.get_index(db)
 

src/genescape/resources.py

@@ -30,6 +30,9 @@ def build_onto_graph(index):
         name = node["name"]
         text = textwrap.fill(name, width=20)
         label = f"{oid}\n{text}"
+        gperc = node.get(utils.GO2GENE_PERC, 0)
+        if gperc > 0.09:
+            label += f": {gperc:.1f}%"
         namespace = utils.NAMESPACE_MAP.get(node["namespace"], "?")
         graph.add_node(oid, id=oid, name=name, namespace=namespace, label=label, **utils.NODE_ATTRS)
 
@@ -209,7 +212,8 @@ def build():
         graph = build_onto_graph(index)
         return graph
 
-    graph = resources.cache("graph", func=build)
+    key = f"graph_{index}"
+    graph = resources.cache(key, func=build)
 
     # Generate the tree from the graph.
     tree = make_tree(ann=ann, graph=graph)

src/genescape/server.py

@@ -89,6 +89,9 @@ def stop(msg):
 # Namespace categories
 NS_BP, NS_MF, NS_CC, NS_ALL = "BP", "MF", "CC", "ALL"
 
+# Additional fields
+GO2GENE_COUNT, GO2GENE_PERC = "gene_count", "gene_percent"
+
 # Map the GO categories namespaces.
 NAMESPACE_MAP = {
     "biological_process": NS_BP,