Add CLO processing notebook (#133)

The [Cell Line Ontology (CLO)](https://bioregistry.io/registry/clo) is a
detailed resource, however it does not follow standard OBO modeling
pattern for cross-references that either a predicate from
[SKOS](https://bioregistry.io/skos) or `oboInOwl:hasDbXref` to point to
a single CURIE encoded as a string. Instead, it uses `rdfs:seeAlso` with
a combination of non-standard CURIEs that are either comma or semi-colon
delimited.

Depends on:
- biopragmatics/bioregistry#896
- CLO-ontology/CLO#103

scripts/generate_clo_mesh_mappings.py

@@ -0,0 +1,38 @@
+# -*- coding: utf-8 -*-
+
+"""Generate mappings to CLO from to MeSH."""
+
+import click
+from more_click import verbose_option
+from semra.sources.clo import get_clo_mappings
+
+from biomappings.gilda_utils import append_gilda_predictions
+from biomappings.mapping_graph import get_filter_from_semra
+from biomappings.utils import get_script_url
+
+
+@click.command()
+@verbose_option
+def main():
+    """Generate CLO-MeSH mappings."""
+    provenance = get_script_url(__file__)
+
+    prefix = "clo"
+    targets = [
+        "mesh",
+        "efo",
+    ]
+
+    clo_mappings = get_clo_mappings()
+    custom_filter = get_filter_from_semra(clo_mappings)
+
+    append_gilda_predictions(
+        prefix,
+        targets,
+        provenance=provenance,
+        custom_filter=custom_filter,
+    )
+
+
+if __name__ == "__main__":
+    main()

src/biomappings/gilda_utils.py

@@ -4,7 +4,7 @@
 
 import logging
 from collections import defaultdict
-from typing import Iterable, Mapping, Optional, Tuple, Union
+from typing import Iterable, Optional, Tuple, Union
 
 import bioregistry
 import pyobo
@@ -13,11 +13,10 @@
 from pyobo.gilda_utils import get_grounder, iter_gilda_prediction_tuples
 
 from biomappings.resources import PredictionTuple, append_prediction_tuples
+from biomappings.utils import CMapping
 
 logger = logging.getLogger(__name__)
 
-CMapping = Mapping[str, Mapping[str, Mapping[str, str]]]
-
 
 def append_gilda_predictions(
     prefix: str,

src/biomappings/mapping_graph.py

@@ -4,15 +4,32 @@
 
 import itertools as itt
 from collections import defaultdict
-from typing import DefaultDict, Dict, Iterable, Mapping, Optional
+from typing import TYPE_CHECKING, DefaultDict, Dict, Iterable, List, Optional
 
 import networkx as nx
 import pyobo
 
+from biomappings.utils import CMapping
 
-def get_custom_filter(
-    prefix: str, targets: Iterable[str]
-) -> Mapping[str, Mapping[str, Mapping[str, str]]]:
+if TYPE_CHECKING:
+    import semra
+
+__all__ = [
+    "get_custom_filter",
+    "get_filter_from_semra",
+    "mutual_mapping_graph",
+]
+
+
+def get_filter_from_semra(mappings: List["semra.Mapping"]) -> CMapping:
+    """Get a custom filter dictionary from a set of SeMRA mappings."""
+    rv: DefaultDict[str, DefaultDict[str, Dict[str, str]]] = defaultdict(lambda: defaultdict(dict))
+    for mapping in mappings:
+        rv[mapping.s.prefix][mapping.o.prefix][mapping.s.identifier] = mapping.o.identifier
+    return rv
+
+
+def get_custom_filter(prefix: str, targets: Iterable[str]) -> CMapping:
     """Get a custom filter dictionary induced over the mutual mapping graph with all target prefixes.
 
     :param prefix: The source prefix

src/biomappings/resources/__init__.py

@@ -125,6 +125,37 @@ def from_dict(cls, mapping: Mapping[str, str]) -> "PredictionTuple":
             values.append(value)
         return cls(*values)  # type:ignore
 
+    @classmethod
+    def from_semra(cls, mapping, confidence) -> "PredictionTuple":
+        """Instantiate from a SeMRA mapping."""
+        import pyobo
+        import semra
+
+        s_name = pyobo.get_name(*mapping.s.pair)
+        if not s_name:
+            raise KeyError(f"could not look up name for {mapping.s.curie}")
+        o_name = pyobo.get_name(*mapping.o.pair)
+        if not o_name:
+            raise KeyError(f"could not look up name for {mapping.o.curie}")
+        # Assume that each mapping has a single simple evidence with a mapping set annotation
+        if len(mapping.evidence) != 1:
+            raise ValueError
+        evidence = mapping.evidence[0]
+        if not isinstance(evidence, semra.SimpleEvidence):
+            raise TypeError
+        if evidence.mapping_set is None:
+            raise ValueError
+        return cls(  # type:ignore
+            *mapping.s.pair,
+            s_name,
+            mapping.p.curie,
+            *mapping.o.pair,
+            o_name,
+            evidence.justification.curie,
+            confidence,
+            evidence.mapping_set.name,
+        )
+
     @property
     def source_curie(self) -> str:
         """Concatenate the source prefix and ID to a CURIE."""
@@ -421,3 +452,20 @@ def get_curated_filter() -> Mapping[str, Mapping[str, Mapping[str, str]]]:
     for m in itt.chain(load_mappings(), load_false_mappings(), load_unsure()):
         d[m["source prefix"]][m["target prefix"]][m["source identifier"]] = m["target identifier"]
     return {k: dict(v) for k, v in d.items()}
+
+
+def prediction_tuples_from_semra(
+    mappings,
+    *,
+    confidence: float,
+) -> List[PredictionTuple]:
+    """Get prediction tuples from SeMRA mappings."""
+    rows = []
+    for mapping in mappings:
+        try:
+            row = PredictionTuple.from_semra(mapping, confidence)
+        except KeyError as e:
+            tqdm.write(str(e))
+            continue
+        rows.append(row)
+    return rows

src/biomappings/resources/incorrect.tsv

@@ -194,6 +194,18 @@ cl	CL:2000004	pituitary gland cell	skos:exactMatch	mesh	D010902	Pituitary Gland
 cl	CL:2000021	sebaceous gland cell	skos:exactMatch	mesh	D012627	Sebaceous Glands	semapv:ManualMappingCuration	orcid:0000-0001-9439-5346			
 cl	CL:2000022	cardiac septum cell	skos:exactMatch	mesh	D006346	Heart Septum	semapv:ManualMappingCuration	orcid:0000-0001-9439-5346			
 cl	CL:2000030	hypothalamus cell	skos:exactMatch	mesh	D007031	Hypothalamus	semapv:ManualMappingCuration	orcid:0000-0001-9439-5346			
+clo	0001922	BE2 cell	skos:exactMatch	mesh	D016175	B-Lymphocyte Subsets	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.502
+clo	0002596	COS-1 cell	skos:exactMatch	mesh	D019556	COS Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0002597	COS-7 cell	skos:exactMatch	mesh	D019556	COS Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0002941	EPI cell	skos:exactMatch	mesh	D015251	Epirubicin	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.556
+clo	0003413	G cell	skos:exactMatch	mesh	D019863	Gastrin-Secreting Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0009279	TC-1 cell	skos:exactMatch	mesh	D013602	T-Lymphocytes, Cytotoxic	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0037163	Ishikawa cell	skos:exactMatch	cellosaurus	CVCL_D199	Ishikawa 3-H-12	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0037237	293-derived cell	skos:exactMatch	cellosaurus	CVCL_0045	HEK293	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0037261	3T3-derived cell	skos:exactMatch	mesh	D016475	3T3 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0037287	COR123 cell	skos:exactMatch	efo	0002142	CORL23	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0051004	RCB0256 cell	skos:exactMatch	cellosaurus	CVCL_1075	BALL-1	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0051005	RCB1882 cell	skos:exactMatch	cellosaurus	CVCL_1075	BALL-1	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
 doid	DOID:0001816	angiosarcoma	skos:exactMatch	mesh	D006394	Hemangiosarcoma	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370			
 doid	DOID:0001816	angiosarcoma	skos:exactMatch	umls	C0018923	Hemangiosarcoma	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370			
 doid	DOID:0001816	angiosarcoma	skos:exactMatch	umls	C0278592	Adult Angiosarcoma	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370			

src/biomappings/resources/mappings.tsv

@@ -3011,6 +3011,26 @@ cl	CL:0008002	skeletal muscle fiber	skos:exactMatch	mesh	D018485	Muscle Fibers,
 cl	CL:0010003	epithelial cell of alveolus of lung	skos:exactMatch	mesh	D056809	Alveolar Epithelial Cells	semapv:ManualMappingCuration	orcid:0000-0001-9439-5346			
 cl	CL:0010017	zygote	skos:exactMatch	mesh	D015053	Zygote	semapv:ManualMappingCuration	orcid:0000-0001-9439-5346			
 cl	CL:0010021	cardiac myoblast	skos:exactMatch	mesh	D032386	Myoblasts, Cardiac	semapv:ManualMappingCuration	orcid:0000-0001-9439-5346			
+clo	0000031	cell line	skos:exactMatch	mesh	D002460	Cell Line	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.762
+clo	0001230	HEK293	skos:exactMatch	cellosaurus	CVCL_0045	HEK293	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0001345	3T3 cell	skos:exactMatch	mesh	D016475	3T3 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0001601	A549 cell	skos:exactMatch	mesh	D000072283	A549 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0002585	COR-L23 cell	skos:exactMatch	efo	0002142	CORL23	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0003704	Hep G2 cell	skos:exactMatch	mesh	D056945	Hep G2 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0007606	MCF7 cell	skos:exactMatch	mesh	D061986	MCF-7 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0007634	MDA-MB-231 cell	skos:exactMatch	efo	0001209	MDAMB231	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0007634	MDA-MB-231 cell	skos:exactMatch	mesh	D000092302	MDA-MB-231 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0007646	MDCK cell	skos:exactMatch	mesh	D061985	Madin Darby Canine Kidney Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0008395	PC-3 cell	skos:exactMatch	mesh	D000078722	PC-3 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0008753	RAW 264.7 cell	skos:exactMatch	mesh	D000067996	RAW 264.7 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0009348	THP-1 cell	skos:exactMatch	mesh	D000074084	THP-1 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0036932	Hybridoma	skos:exactMatch	mesh	D006825	Hybridomas	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.556
+clo	0036936	Somatic cell hybrid	skos:exactMatch	mesh	D006822	Hybrid Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0037116	LNCaP cell	skos:exactMatch	efo	0002071	LNCAP	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/b250ea/scripts/generate_clo_mesh_mappings.py	0.53
+clo	0037230	Ishikawa 3-H-12 cell	skos:exactMatch	cellosaurus	CVCL_D199	Ishikawa 3-H-12	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0037291	MDAMB231 cell	skos:exactMatch	mesh	D000092302	MDA-MB-231 Cells	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.549
+clo	0037300	BALL-1 cell	skos:exactMatch	cellosaurus	CVCL_1075	BALL-1	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:UnspecifiedMatching	clo	0.8
+clo	0037339	tissue donor	skos:exactMatch	mesh	D014019	Tissue Donors	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370	semapv:LexicalMatching	https://github.com/biomappings/biomappings/blob/746fde/scripts/generate_clo_mesh_mappings.py	0.54
 doid	DOID:0040002	aspirin allergy	skos:exactMatch	umls	C0004058	Allergy to aspirin	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370			
 doid	DOID:0040004	amoxicillin allergy	skos:exactMatch	umls	C0571417	Allergy to amoxicillin	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370			
 doid	DOID:0040005	ceftriaxone allergy	skos:exactMatch	umls	C0571463	Allergy to ceftriaxone	semapv:ManualMappingCuration	orcid:0000-0003-4423-4370