Add neo4j nomencalture mapping

hetnet/neo4j/README.md

@@ -2,6 +2,10 @@
 
 `hetionet-v1.0.db.tar.bz2` is a bzip2-compressed tarball of a Neo4j database. The database was build under Neo4j 2.3.3 Community Edition.
 
+## Nomenclature
+
+The Neo4j graph uses slightly different names for metanodes (Neo4j labels) and metaedges (Neo4j relationship types). See [`labels.tsv`](labels.tsv) for the node type mapping and [`types.tsv`](types.tsv) for the edge type mapping. We conform to the Neo4j style of CamelCase labels and ALL_CAPS relationship types. In addition, Neo4j relationship types are appended with metaedge standard abbreviations. This adds source/target metanode awareness to relationship types and [enables](https://thinklab.com/discussion/using-the-neo4j-graph-database-for-hetnets/112#6 "Query Optimization · Using the neo4j graph database for hetnets · Thinklab Discussion in Project Rephetio") optimized queries.
+
 ## Neo4j setup instructions
 
 Download and extract Neo4j 2.3 Community Edition [from Neo4j](http://neo4j.com/download/other-releases/ "Neo4j Releases"). Navigate to the `data` directory inside the Neo4j directory and run:

hetnet/neo4j/labels.tsv

@@ -0,0 +1,12 @@
+metanode	abbreviation	label
+Anatomy	A	Anatomy
+Biological Process	BP	BiologicalProcess
+Cellular Component	CC	CellularComponent
+Compound	C	Compound
+Disease	D	Disease
+Gene	G	Gene
+Molecular Function	MF	MolecularFunction
+Pathway	PW	Pathway
+Pharmacologic Class	PC	PharmacologicClass
+Side Effect	SE	SideEffect
+Symptom	S	Symptom

hetnet/neo4j/types.tsv

@@ -0,0 +1,25 @@
+metaedge	rel_type	direction
+Anatomy - downregulates - Gene	DOWNREGULATES_AdG	both
+Anatomy - expresses - Gene	EXPRESSES_AeG	both
+Anatomy - upregulates - Gene	UPREGULATES_AuG	both
+Compound - binds - Gene	BINDS_CbG	both
+Compound - causes - Side Effect	CAUSES_CcSE	both
+Compound - downregulates - Gene	DOWNREGULATES_CdG	both
+Compound - palliates - Disease	PALLIATES_CpD	both
+Compound - resembles - Compound	RESEMBLES_CrC	both
+Compound - treats - Disease	TREATS_CtD	both
+Compound - upregulates - Gene	UPREGULATES_CuG	both
+Disease - associates - Gene	ASSOCIATES_DaG	both
+Disease - downregulates - Gene	DOWNREGULATES_DdG	both
+Disease - localizes - Anatomy	LOCALIZES_DlA	both
+Disease - presents - Symptom	PRESENTS_DpS	both
+Disease - resembles - Disease	RESEMBLES_DrD	both
+Disease - upregulates - Gene	UPREGULATES_DuG	both
+Gene - covaries - Gene	COVARIES_GcG	both
+Gene - interacts - Gene	INTERACTS_GiG	both
+Gene - participates - Biological Process	PARTICIPATES_GpBP	both
+Gene - participates - Cellular Component	PARTICIPATES_GpCC	both
+Gene - participates - Molecular Function	PARTICIPATES_GpMF	both
+Gene - participates - Pathway	PARTICIPATES_GpPW	both
+Gene > regulates > Gene	REGULATES_GrG	forward
+Pharmacologic Class - includes - Compound	INCLUDES_PCiC	both

import.ipynb

@@ -32,7 +32,7 @@
    "outputs": [],
    "source": [
     "# dhimmel/integrate commit\n",
-    "commit = '470bd40e6ac763f7690f55818ed21afbaf73163b'\n",
+    "commit = 'ffd1a48b4051c41fc8cef6e8847d0687f1a722bc'\n",
     "\n",
     "# Name and version for hetionet\n",
     "name = 'hetionet-v1.0'"
@@ -68,6 +68,10 @@
     "    'data/summary/metaedge-styles.tsv': 'describe/edges/metaedge-styles.tsv',\n",
     "    'data/summary/degrees.xlsx': 'describe/degree/degrees.xlsx',\n",
     "    'viz/degrees.pdf': 'describe/degree/degrees.pdf',\n",
+    "\n",
+    "    # Neo4j nomencalture mappings\n",
+    "    'neo4j/nomenclature/labels.tsv': 'hetnet/neo4j/labels.tsv',\n",
+    "    'neo4j/nomenclature/types.tsv': 'hetnet/neo4j/types.tsv',\n",
     "}"
    ]
   },
@@ -81,8 +85,8 @@
    "source": [
     "for integrate_path, hetionet_path in integrate_to_hetionet.items():\n",
     "    url = 'https://github.com/dhimmel/integrate/raw/{}/{}'.format(commit, integrate_path)\n",
-    "#     if os.path.exists(hetionet_path):\n",
-    "#         continue\n",
+    "    if os.path.exists(hetionet_path):\n",
+    "        continue\n",
     "    urllib.request.urlretrieve(url, filename=hetionet_path)"
    ]
   },