Update with new SeMRA API (#17)

Author: cthoyt

lexica/anatomy/generate.py

@@ -33,8 +33,6 @@
 GILDA_PATH = HERE.joinpath("terms.tsv.gz")
 SUMMARY_PATH = HERE.joinpath("summary.json")
 
-ANATOMY_CONFIGURATION.add_mapping_caches(HERE)
-
 
 @click.command()
 def _main() -> None:

lexica/cell/generate.py

@@ -31,8 +31,6 @@
 GILDA_PATH = HERE.joinpath("terms.tsv.gz")
 SUMMARY_PATH = HERE.joinpath("summary.json")
 
-CELL_CONFIGURATION.add_mapping_caches(HERE)
-
 
 def _main() -> None:
     """Generate a lexical index for cell resources."""

lexica/phenotype/generate.py

@@ -31,8 +31,6 @@
 GILDA_PATH = HERE.joinpath("terms.tsv.gz")
 SUMMARY_PATH = HERE.joinpath("summary.json")
 
-PHENOTYPE_CONFIGURATION.add_mapping_caches(HERE)
-
 
 def _main() -> None:
     """Generate a lexical index for phenotype resources."""

pyproject.toml

@@ -59,7 +59,7 @@ dependencies = [
     "semra",
     "ssslm>=0.0.10",
     "bioregistry",
-    "pyobo @ git+https://github.com/biopragmatics/pyobo",
+    "pyobo",
     "bioontologies>=0.6.0",
     "biosynonyms",
     "pandas",

src/biolexica/api.py

@@ -57,18 +57,6 @@ class Configuration(BaseModel):
     )
     mapping_configuration: semra.Configuration | None = None
 
-    def add_mapping_caches(self, directory: Path) -> None:
-        """Add a cache to the directory for raw, processed, and priority mappings."""
-        if self.mapping_configuration is None:
-            raise ValueError
-        self.mapping_configuration.raw_pickle_path = directory.joinpath("mappings_raw.pkl.gz")
-        self.mapping_configuration.processed_pickle_path = directory.joinpath(
-            "mappings_processed.pkl.gz"
-        )
-        self.mapping_configuration.priority_pickle_path = directory.joinpath(
-            "mappings_prioritized.pkl"
-        )
-
 
 PREDEFINED: TypeAlias = Literal["cell", "anatomy", "phenotype", "obo"]
 URL_FMT = "https://github.com/biopragmatics/biolexica/raw/main/lexica/{key}/{key}.ssslm.tsv.gz"
@@ -146,9 +134,15 @@ def assemble_terms(  # noqa:C901
         logger.info("Writing %d raw literal mappings to %s", len(terms), raw_path)
         ssslm.write_literal_mappings(terms, raw_path)
 
-    _mappings = []
-    if configuration.mapping_configuration:
-        _mappings.extend(configuration.mapping_configuration.get_mappings())
+    _mappings: list[semra.Mapping] = []
+    if configuration.mapping_configuration is not None:
+        from semra.pipeline import AssembleReturnType
+
+        _mappings.extend(
+            configuration.mapping_configuration.get_mappings(
+                return_type=AssembleReturnType.priority
+            )
+        )
     if mappings is not None:
         _mappings.extend(mappings)
 
@@ -158,7 +152,7 @@ def assemble_terms(  # noqa:C901
         assert_projection(_mappings)
         terms = ssslm.remap_literal_mappings(
             literal_mappings=terms,
-            mappings=[(mapping.s, mapping.o) for mapping in _mappings],
+            mappings=[(mapping.subject, mapping.object) for mapping in _mappings],
         )
 
     if configuration.excludes:

src/biolexica/configs/anatomy.py

@@ -1,46 +1,12 @@
 """Configuration for an anatomy lexical index."""
 
-import semra
+import semra.landscape.anatomy
 from pyobo.sources.mesh import get_mesh_category_curies
 
 import biolexica
 
 __all__ = ["ANATOMY_CONFIGURATION"]
 
-PRIORITY = [
-    "uberon",
-    "mesh",
-    "bto",
-    "caro",
-    "ncit",
-    "umls",
-]
-
-SEMRA_CONFIG = semra.Configuration(
-    name="Anatomy mappings",
-    inputs=[
-        semra.Input(source="biomappings"),
-        semra.Input(source="gilda"),
-        semra.Input(prefix="uberon", source="pyobo", confidence=0.99),
-        semra.Input(prefix="bto", source="pyobo", confidence=0.99),
-        semra.Input(prefix="caro", source="pyobo", confidence=0.99),
-        semra.Input(prefix="mesh", source="pyobo", confidence=0.99),
-        semra.Input(prefix="ncit", source="pyobo", confidence=0.99),
-        semra.Input(prefix="umls", source="pyobo", confidence=0.99),
-    ],
-    add_labels=False,
-    priority=PRIORITY,
-    keep_prefixes=PRIORITY,
-    remove_imprecise=False,
-    mutations=[
-        semra.Mutation(source="uberon", confidence=0.8),
-        semra.Mutation(source="bto", confidence=0.65),
-        semra.Mutation(source="caro", confidence=0.8),
-        semra.Mutation(source="ncit", confidence=0.7),
-        semra.Mutation(source="umls", confidence=0.7),
-    ],
-)
-
 ANATOMY_CONFIGURATION = biolexica.Configuration(
     inputs=[
         biolexica.Input(source="uberon", processor="pyobo"),
@@ -63,5 +29,5 @@
             source="umls", processor="pyobo", ancestors=["umls:C0700276", "umls:C1515976"]
         ),
     ],
-    mapping_configuration=SEMRA_CONFIG,
+    mapping_configuration=semra.landscape.anatomy.ANATOMY_CONFIGURATION,
 )

src/biolexica/configs/cell.py

@@ -1,69 +1,12 @@
 """Configuration for a cell lexical index."""
 
 import semra
+import semra.landscape.cell
 
 import biolexica
 
 __all__ = ["CELL_CONFIGURATION"]
 
-
-PRIORITY = [
-    "cl",
-    "cellosaurus",
-    "bto",
-    "clo",
-    "efo",
-    "mesh",
-    "ccle",
-    "depmap",
-]
-
-
-SEMRA_CONFIG = semra.Configuration(
-    name="Cell and Cell Line Mappings",
-    description="Originally a reproduction of the EFO/Cellosaurus/DepMap/CCLE scenario "
-    "posed in the Biomappings paper, this configuration imports several different cell and "
-    "cell line resources and identifies mappings between them.",
-    inputs=[
-        semra.Input(source="biomappings"),
-        semra.Input(source="gilda"),
-        semra.Input(prefix="cellosaurus", source="pyobo", confidence=0.99),
-        semra.Input(prefix="bto", source="bioontologies", confidence=0.99),
-        semra.Input(prefix="cl", source="bioontologies", confidence=0.99),
-        semra.Input(prefix="clo", source="custom", confidence=0.99),
-        semra.Input(prefix="efo", source="pyobo", confidence=0.99),
-        semra.Input(
-            prefix="depmap",
-            source="pyobo",
-            confidence=0.99,
-            extras={"version": "22Q4", "standardize": True, "license": "CC-BY-4.0"},
-        ),
-        semra.Input(
-            prefix="ccle",
-            source="pyobo",
-            confidence=0.99,
-            extras={"version": "2019"},
-        ),
-        semra.Input(prefix="ncit", source="pyobo", confidence=0.99),
-        semra.Input(prefix="umls", source="pyobo", confidence=0.99),
-    ],
-    add_labels=False,
-    priority=PRIORITY,
-    keep_prefixes=PRIORITY,
-    remove_imprecise=False,
-    mutations=[
-        semra.Mutation(source="efo", confidence=0.7),
-        semra.Mutation(source="bto", confidence=0.7),
-        semra.Mutation(source="cl", confidence=0.7),
-        semra.Mutation(source="clo", confidence=0.7),
-        semra.Mutation(source="depmap", confidence=0.7),
-        semra.Mutation(source="ccle", confidence=0.7),
-        semra.Mutation(source="cellosaurus", confidence=0.7),
-        semra.Mutation(source="ncit", confidence=0.7),
-        semra.Mutation(source="umls", confidence=0.7),
-    ],
-)
-
 CELL_CONFIGURATION = biolexica.Configuration(
     inputs=[
         biolexica.Input(
@@ -87,5 +30,5 @@
             ],
         ),
     ],
-    mapping_configuration=SEMRA_CONFIG,
+    mapping_configuration=semra.landscape.cell.CELL_CONFIGURATION,
 )

src/biolexica/configs/phenotype.py

@@ -1,52 +1,12 @@
 """Configuration for a phenotype lexical index."""
 
-import semra
+import semra.landscape.disease
 from pyobo.sources.mesh import get_mesh_category_curies
 
 import biolexica
 
 __all__ = ["PHENOTYPE_CONFIGURATION"]
 
-PRIORITY = [
-    "doid",
-    "mondo",
-    "hp",
-    "symp",
-    "mesh",
-    "efo",
-]
-
-SEMRA_CONFIG = semra.Configuration(
-    name="Cell and Cell Line Mappings",
-    description="Originally a reproduction of the EFO/Cellosaurus/DepMap/CCLE scenario "
-    "posed in the Biomappings paper, this configuration imports several different cell and "
-    "cell line resources and identifies mappings between them.",
-    inputs=[
-        semra.Input(source="biomappings"),
-        semra.Input(source="gilda"),
-        semra.Input(prefix="doid", source="pyobo", confidence=0.99),
-        semra.Input(prefix="mondo", source="pyobo", confidence=0.99),
-        semra.Input(prefix="hp", source="pyobo", confidence=0.99),
-        semra.Input(prefix="symp", source="pyobo", confidence=0.99),
-        semra.Input(prefix="mesh", source="pyobo", confidence=0.99),
-        semra.Input(prefix="efo", source="pyobo", confidence=0.99),
-        semra.Input(prefix="umls", source="pyobo", confidence=0.99),
-        semra.Input(prefix="ncit", source="pyobo", confidence=0.99),
-    ],
-    add_labels=False,
-    priority=PRIORITY,
-    keep_prefixes=PRIORITY,
-    remove_imprecise=False,
-    mutations=[
-        semra.Mutation(source="doid", confidence=0.7),
-        semra.Mutation(source="mondo", confidence=0.7),
-        semra.Mutation(source="hp", confidence=0.7),
-        semra.Mutation(source="symp", confidence=0.7),
-        semra.Mutation(source="umls", confidence=0.7),
-        semra.Mutation(source="ncit", confidence=0.7),
-    ],
-)
-
 PHENOTYPE_CONFIGURATION = biolexica.Configuration(
     inputs=[
         biolexica.Input(source="doid", processor="pyobo"),
@@ -67,5 +27,5 @@
         biolexica.Input(source="umls", processor="pyobo", ancestors=["umls:C0012634"]),
     ],
     excludes=["doid:4"],
-    mapping_configuration=SEMRA_CONFIG,
+    mapping_configuration=semra.landscape.disease.DISEASE_CONFIGURATION,
 )