Merge pull request #35 from INCATools/gap-filling

gap filling

.gitignore

@@ -5,3 +5,12 @@ __pycache__
 docs/_build/
 docs/datamodels/*/*.md
 tests/output/
+
+dist/
+db/
+
+notebooks/output/*json
+notebooks/output/*tsv
+notebooks/input/go.db
+
+

notebooks/Command-Line-Examples.ipynb

@@ -105,6 +105,51 @@
     "The default ubergraph endpoint is used, but you could also set up your own"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "8dbaac3c",
+   "metadata": {},
+   "source": [
+    "### Graphy operations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "ca33e9c1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CL:0000000 ! cell\r\n",
+      "CL:0000540 ! neuron\r\n",
+      "CL:0002319 ! neural cell\r\n",
+      "GO:0001750 ! photoreceptor outer segment\r\n",
+      "GO:0005575 ! cellular_component\r\n",
+      "UBERON:0001062 ! anatomical entity\r\n",
+      "BFO:0000040 ! material entity\r\n",
+      "CARO:0000000 ! anatomical entity\r\n",
+      "CL:0000003 ! native cell\r\n",
+      "CL:0000211 ! electrically active cell\r\n",
+      "CL:0000255 ! eukaryotic cell\r\n",
+      "CL:0000393 ! electrically responsive cell\r\n",
+      "CL:0000404 ! electrically signaling cell\r\n",
+      "CL:0000548 ! animal cell\r\n",
+      "CL:0002371 ! somatic cell\r\n",
+      "BFO:0000002 ! continuant\r\n",
+      "BFO:0000004 ! independent continuant\r\n",
+      "CARO:0030000 ! biological entity\r\n",
+      "GO:0110165 ! cellular anatomical entity\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "# all is-a (i.e. subClassOf between named classes) ancestors of a seed set\n",
+    "!runoak  -i ubergraph: ancestors GO:0001750 CL:0000540 -p i"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 40,
@@ -121,6 +166,7 @@
     }
    ],
    "source": [
+    "# visualize starting from a seed set of terms\n",
     "!runoak  -i ubergraph: viz --no-view GO:0001750 CL:0000540 -p i,BFO:0000050 -o output/ug-test1.png -O png"
    ]
   },
@@ -132,6 +178,41 @@
     "![img](output/ug-test1.png)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "70dea17b",
+   "metadata": {},
+   "source": [
+    "### Gap Filling"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "1a9d16e8",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(node:2377) [DEP0128] DeprecationWarning: Invalid 'main' field in '/Users/cjm/repos/obographviz/node_modules/node-getopt/package.json' of './lib'. Please either fix that or report it to the module author\r\n",
+      "(Use `node --trace-deprecation ...` to show where the warning was created)\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!runoak -i ubergraph: viz --gap-fill GO:0001750 CL:0000540 CL:0000000 NCBITaxon:9606 NCBITaxon:1 -p i,p -o output/ug-gap-fill1.png -O png"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "695bb837",
+   "metadata": {},
+   "source": [
+    "![img](output/ug-gap-fill1.png)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "24d6a47e",

src/oaklib/cli.py

@@ -250,6 +250,10 @@ def annotate(words, output: str):
               default=False,
               show_default=True,
               help="traverse down")
+@click.option("--gap-fill/--no-gap-fill",
+              default=False,
+              show_default=True,
+              help="If set then find the minimal graph that spans all input curies")
 @click.option('-S', '--stylemap',
               help='a json file to configure visualization. See https://berkeleybop.github.io/kgviz-model/')
 @click.option('-C', '--configure',
@@ -261,7 +265,7 @@ def annotate(words, output: str):
 @click.option('-o', '--output',
               help="Path to output file")
 #@output_option
-def viz(terms, predicates, down, view, stylemap, configure, output_type: str, output: str):
+def viz(terms, predicates, down, gap_fill, view, stylemap, configure, output_type: str, output: str):
     """
     Visualizing an ancestor graph using obographviz
 
@@ -308,17 +312,27 @@ def viz(terms, predicates, down, view, stylemap, configure, output_type: str, ou
             curies = terms_expanded
         if down:
             graph = impl.subgraph(curies, predicates=actual_predicates)
+        elif gap_fill:
+            logging.info(f'Using gap-fill strategy')
+            if isinstance(impl, SubsetterInterface):
+                rels = impl.gap_fill_relationships(curies, predicates=actual_predicates)
+                if isinstance(impl, OboGraphInterface):
+                    graph = impl.relationships_to_graph(rels)
+                else:
+                    assert False
+            else:
+                raise NotImplementedError(f'{impl} needs to implement Subsetter for --gap-fill')
         else:
             graph = impl.ancestor_graph(curies, predicates=actual_predicates)
         logging.info(f'Drawing graph seeded from {curies}')
         if output_type == 'json':
             if output:
-                json_dumper.dump(graph, to_file=output)
+                json_dumper.dump(graph, to_file=output, inject_type=False)
             else:
                 print(json_dumper.dumps(graph))
         elif output_type == 'yaml':
             if output:
-                yaml_dumper.dump(graph, to_file=output)
+                yaml_dumper.dump(graph, to_file=output, inject_type=False)
             else:
                 print(yaml_dumper.dumps(graph))
         else:
@@ -369,6 +383,7 @@ def ancestors(terms, predicates, output: str):
     else:
         raise NotImplementedError(f'Cannot execute this using {impl} of type {type(impl)}')
 
+
 @main.command()
 @click.argument("terms", nargs=-1)
 @predicates_option

src/oaklib/implementations/sqldb/sql_implementation.py

@@ -2,15 +2,16 @@
 from abc import ABC
 from collections import defaultdict
 from dataclasses import dataclass
-from typing import List, Any, Iterable, Optional, Type, Dict, Union, Tuple
+from typing import List, Any, Iterable, Optional, Type, Dict, Union, Tuple, Iterator
 
 from linkml_runtime import SchemaView
 from linkml_runtime.utils.introspection import package_schemaview
 from linkml_runtime.utils.metamodelcore import URIorCURIE
 from oaklib.implementations.sqldb.model import Statements, Edge, HasSynonymStatement, \
     HasTextDefinitionStatement, ClassNode, IriNode, RdfsLabelStatement, DeprecatedNode, EntailedEdge, \
     ObjectPropertyNode, AnnotationPropertyNode, NamedIndividualNode, HasMappingStatement
-from oaklib.interfaces.basic_ontology_interface import RELATIONSHIP_MAP, PRED_CURIE, ALIAS_MAP
+from oaklib.interfaces import SubsetterInterface
+from oaklib.interfaces.basic_ontology_interface import RELATIONSHIP_MAP, PRED_CURIE, ALIAS_MAP, RELATIONSHIP
 from oaklib.interfaces.mapping_provider_interface import MappingProviderInterface
 from oaklib.interfaces.obograph_interface import OboGraphInterface
 from oaklib.interfaces.relation_graph_interface import RelationGraphInterface
@@ -20,6 +21,7 @@
 from oaklib.datamodels import obograph, ontology_metadata
 import oaklib.datamodels.validation_datamodel as vdm
 from oaklib.datamodels.vocabulary import SYNONYM_PREDICATES, omd_slots, LABEL_PREDICATE, IN_SUBSET
+from oaklib.utilities.graph.networkx_bridge import transitive_reduction_by_predicate
 from sqlalchemy import select, text, exists
 from sqlalchemy.orm import sessionmaker, aliased
 from sqlalchemy import create_engine
@@ -38,7 +40,8 @@ def get_range_xsd_type(sv: SchemaView, rng: str) -> Optional[URIorCURIE]:
 
 
 @dataclass
-class SqlImplementation(RelationGraphInterface, OboGraphInterface, ValidatorInterface, SearchInterface, MappingProviderInterface, ABC):
+class SqlImplementation(RelationGraphInterface, OboGraphInterface, ValidatorInterface, SearchInterface,
+                        SubsetterInterface, MappingProviderInterface, ABC):
     """
     A :class:`OntologyInterface` implementation that wraps a SQL Relational Database
 
@@ -381,5 +384,15 @@ def _check_slot(self, slot_name: str, class_name: str = 'Class') -> Iterable[vdm
                                                       )
                         yield result
 
-
-
+    def gap_fill_relationships(self, seed_curies: List[CURIE], predicates: List[PRED_CURIE] = None) -> Iterator[RELATIONSHIP]:
+        seed_curies = tuple(seed_curies)
+        q = self.session.query(EntailedEdge).filter(EntailedEdge.subject.in_(seed_curies))
+        q = q.filter(EntailedEdge.object.in_(seed_curies))
+        if predicates:
+            q = q.filter(EntailedEdge.predicate.in_(tuple(predicates)))
+        rels = []
+        for row in q:
+            if row.subject != row.object:
+                rels.append((row.subject, row.predicate, row.object))
+        for rel in transitive_reduction_by_predicate(rels):
+            yield rel