Add first version of enhanced mapping metadata

.claude/settings.json

@@ -1,7 +1,7 @@
 {
   "permissions": {
     "allow": [
-        "Bash(*)",
+        "Bash",
         "Edit",
         "MultiEdit",
         "NotebookEdit",

project.Makefile

@@ -13,3 +13,48 @@ docs/specification/compliance.md: tests/test_compliance/test_compliance_suite.py
 doctest:
 	$(RUN) python -m doctest --option ELLIPSIS --option NORMALIZE_WHITESPACE src/linkml_map/*.py src/linkml_map/*/*.py
 
+# Transform a LinkML-Map TransformationSpecification to FAIR Mappings Schema
+# This demonstrates mapping metadata elements from linkml-map to fair-mappings-schema
+FAIR_EXAMPLE_DIR = tests/input/examples/fair_mappings_metadata
+FAIR_OUTPUT_DIR = $(FAIR_EXAMPLE_DIR)/output
+
+FAIR_LINKMLMAP_SPEC = $(FAIR_EXAMPLE_DIR)/transform/linkmlmap-to-fair.transformation.yaml
+LINKML_SAMPLE_INPUT = $(FAIR_EXAMPLE_DIR)/data/sample-linkmlmap-spec.yaml
+LINKMLMAP_SCHEMA = src/linkml_map/datamodel/transformer_model.yaml
+
+$(FAIR_OUTPUT_DIR)/linkmlmap-fair-mappings.yaml: $(LINKML_SAMPLE_INPUT) $(FAIR_LINKMLMAP_SPEC)
+	mkdir -p $(FAIR_OUTPUT_DIR)
+	$(RUN) linkml-map map-data \
+		-T $(FAIR_LINKMLMAP_SPEC) \
+		-s $(LINKMLMAP_SCHEMA) \
+		--source-type TransformationSpecification \
+		--unrestricted-eval \
+		$(LINKML_SAMPLE_INPUT) \
+		-o $@
+
+transform-to-fair: $(FAIR_OUTPUT_DIR)/linkmlmap-fair-mappings.yaml
+	@echo "Transformed $(LINKML_SAMPLE_INPUT) to FAIR Mappings Schema format"
+	@echo "Output: $(FAIR_OUTPUT_DIR)/linkmlmap-fair-mappings.yaml"
+
+# Transform an SSSOM Mapping Set to FAIR Mappings Schema
+# This demonstrates mapping SSSOM metadata to fair-mappings-schema
+SSSOM_TRANSFORM_SPEC = $(FAIR_EXAMPLE_DIR)/transform/sssom-to-fair.transformation.yaml
+SSSOM_SAMPLE_INPUT = $(FAIR_EXAMPLE_DIR)/data/sample-sssom-mapping-set.yaml
+SSSOM_SCHEMA = /Users/matentzn/ws/SSSOM/src/sssom_schema/schema/sssom_schema.yaml
+
+$(FAIR_OUTPUT_DIR)/sssom-fair-mappings.yaml: $(SSSOM_SAMPLE_INPUT) $(SSSOM_TRANSFORM_SPEC)
+	mkdir -p $(FAIR_OUTPUT_DIR)
+	$(RUN) linkml-map map-data \
+		-T $(SSSOM_TRANSFORM_SPEC) \
+		-s $(SSSOM_SCHEMA) \
+		--source-type 'mapping set' \
+		--unrestricted-eval \
+		$(SSSOM_SAMPLE_INPUT) \
+		-o $@
+
+sssom-to-fair: $(FAIR_OUTPUT_DIR)/sssom-fair-mappings.yaml
+	@echo "Transformed $(SSSOM_SAMPLE_INPUT) to FAIR Mappings Schema format"
+	@echo "Output: $(FAIR_OUTPUT_DIR)/sssom-fair-mappings.yaml"
+
+# Run all FAIR mappings transformations
+all-fair-transforms: transform-to-fair sssom-to-fair

src/linkml_map/datamodel/transformer_model.yaml

@@ -23,6 +23,7 @@ prefixes:
   rdfs: http://www.w3.org/2000/01/rdf-schema#
   sh: http://www.w3.org/ns/shacl#
   STATO: http://purl.obolibrary.org/obo/STATO_
+  sssom: https://w3id.org/sssom/
 default_prefix: linkmlmap
 
 imports:
@@ -73,8 +74,21 @@ classes:
         description: Unique identifier for this transformation specification
         slot_uri: schema:identifier
       title:
+        # Corresponds to fair_mappings_schema:MappingSpecification.name
         description: human readable title for this transformation specification
         slot_uri: dcterms:title
+      publication_date:
+        description: date of publication of this transformation specification
+        range: date
+        slot_uri: dcterms:issued
+      license:
+        description: license under which this transformation specification is published
+        range: uriorcurie
+        slot_uri: dcterms:license
+      version:
+        description: version of this transformation specification
+        range: string
+        slot_uri: dcterms:version
       prefixes:
         description: maps prefixes to URL expansions
         range: KeyVal
@@ -86,15 +100,47 @@ classes:
         multivalued: true
         inlined: true
       source_schema:
+        # Corresponds to fair_mappings_schema:MappingSpecification.subject_source
         description: name of the schema that describes the source (input) objects
       target_schema:
+        # Corresponds to fair_mappings_schema:MappingSpecification.object_source
         description: name of the schema that describes the target (output) objects
       source_schema_patches:
         range: Any
         description: >-
           Schema patches to apply to the source schema before transformation.
           Useful for adding foreign key relationships to auto-generated schemas.
           Uses LinkML schema YAML structure (classes, slots, attributes, etc.).
+      creator:
+        description: A list of creators of this transformation specification
+        range: Agent
+        multivalued: true
+        inlined: true
+        inlined_as_list: true
+        slot_uri: dcterms:creator
+      author:
+        description: A list of authors of this transformation specification
+        range: Agent
+        multivalued: true
+        inlined: true
+        inlined_as_list: true
+        slot_uri: dcterms:contributor
+      reviewer:
+        description: A list of reviewers of this transformation specification
+        range: Agent
+        multivalued: true
+        inlined: true
+        inlined_as_list: true
+      mapping_method:
+        description: The method used to create this mapping, e.g. manual curation, automated
+          mapping, etc.
+        range: uriorcurie
+      documentation:
+        description: URL or reference to documentation for the mapping specification
+        range: uri
+      content_url:
+        description: Reference to the actual content of the mapping specification
+        range: uri
       class_derivations:
         description: >-
           Instructions on how to derive a set of classes in the target schema
@@ -458,6 +504,60 @@ classes:
       value:
         range: Any
 
+  # Aligned with fair_mappings_schema:Agent
+  Agent:
+    abstract: true
+    description: An entity that can create or contribute to a digital object, such as an author or creator.
+    attributes:
+      id:
+        identifier: true
+        description: Identifier for the agent
+        range: uriorcurie
+      name:
+        description: Name of the agent
+        range: string
+        slot_uri: schema:name
+      type:
+        description: Type of the agent
+        range: string
+        designates_type: true
+
+  # Aligned with fair_mappings_schema:Person
+  Person:
+    is_a: Agent
+    description: An individual person who contributes to a mapping specification
+    attributes:
+      orcid:
+        description: ORCID identifier for the person
+        range: uriorcurie
+      affiliation:
+        description: Institutional affiliation of the person
+        range: string
+
+  # Aligned with fair_mappings_schema:Organization
+  Organization:
+    is_a: Agent
+    description: An organization or institution that contributes to a mapping specification
+    attributes:
+      ror_id:
+        description: ROR (Research Organization Registry) identifier
+        range: uriorcurie
+      url:
+        description: URL or web address of the organization
+        range: uri
+
+  # Aligned with fair_mappings_schema:Software
+  Software:
+    is_a: Agent
+    description: A software tool or system used in creating mappings
+    attributes:
+      version:
+        description: Version of the software
+        range: string
+      repository_url:
+        description: URL to a code repository
+        range: uri
+
   CopyDirective:
     description: >-
       Instructs a Schema Mapper in how to map to a target schema. Not used for data transformation.

tests/input/examples/fair_mappings_metadata/data/sample-linkmlmap-spec.yaml

@@ -0,0 +1,76 @@
+# Sample TransformationSpecification instance for testing transformation to FAIR Mappings Schema
+# This example includes both metadata AND actual transformation content.
+# When transformed to FAIR Mappings Schema, only the metadata elements are extracted;
+# the class_derivations, enum_derivations etc. are linkml-map specific and not part of FAIR Mappings.
+#
+# NOTE: class_derivations and enum_derivations are commented out below due to a bug
+# in dynamic_object.py that doesn't handle key-based (vs identifier-based) classes.
+# The actual transformation would ignore these anyway since FAIR Mappings Schema
+# doesn't have equivalents for these linkml-map specific elements.
+
+id: https://example.org/transformations/gene-expression-mapping
+title: Gene Expression Data Transformation
+description: |-
+  A transformation specification that maps gene expression data from
+  a legacy format to a standardized biomedical data model.
+
+  This specification includes class derivations for:
+  - GeneExpressionMixin (from GeneExpressionRecord)
+  - OrganismTaxon (from SampleMetadata)
+  - Study (from ExperimentInfo)
+
+  And enum derivations for:
+  - ExpressionLevelCategory (from LegacyExpressionLevel)
+publication_date: "2024-01-15"
+license: https://creativecommons.org/licenses/by/4.0/
+version: "1.0.0"
+mapping_method: semapv:ManualMappingCuration
+documentation: https://example.org/docs/gene-expression-mapping
+content_url: https://example.org/content/gene-expression-mapping.yaml
+
+# Using Person type for creator/author/reviewer
+creator:
+  - type: Person
+    id: orcid:0000-0002-1234-5678
+    name: Jane Doe
+
+author:
+  - type: Person
+    id: orcid:0000-0003-9876-5432
+    name: John Smith
+
+reviewer:
+  - type: Person
+    id: orcid:0000-0001-1111-2222
+    name: Alice Johnson
+
+# source_schema and target_schema (simple strings in linkml-map)
+source_schema: legacy_gene_expression.yaml
+target_schema: biolink_model.yaml
+
+# ============================================================================
+# ACTUAL TRANSFORMATION CONTENT (commented out due to dynamic_object bug)
+# These elements are linkml-map specific and won't appear in FAIR Mappings output
+# ============================================================================
+#
+# class_derivations:
+#   GeneExpressionMixin:
+#     populated_from: GeneExpressionRecord
+#     slot_derivations:
+#       id:
+#         populated_from: record_id
+#       gene_id:
+#         populated_from: gene_symbol
+#
+#   OrganismTaxon:
+#     populated_from: SampleMetadata
+#     slot_derivations:
+#       id:
+#         populated_from: taxon_id
+#
+# enum_derivations:
+#   ExpressionLevelCategory:
+#     populated_from: LegacyExpressionLevel
+#     permissible_value_derivations:
+#       HIGH:
+#         populated_from: H

tests/input/examples/fair_mappings_metadata/data/sample-sssom-mapping-set.yaml

@@ -0,0 +1,69 @@
+# Sample SSSOM Mapping Set instance for testing transformation to FAIR Mappings Schema
+# This demonstrates how SSSOM metadata maps to FAIR Mappings Schema MappingSpecification
+mapping_set_id: http://purl.obolibrary.org/obo/mondo/mappings/mondo_exactmatch_ncit.sssom.tsv
+mapping_set_title: Mondo to NCIT Exact Mappings
+mapping_set_description: |-
+  This mapping set contains exact mappings between Mondo Disease Ontology
+  and NCI Thesaurus (NCIT) terms. These mappings were curated by the
+  Monarch Initiative team to support disease data integration.
+mapping_set_version: "2024-01-15"
+license: https://creativecommons.org/publicdomain/zero/1.0/
+publication_date: "2024-01-15"
+
+# Creator and author information (SSSOM uses ID + label pattern)
+creator_id:
+  - orcid:0000-0002-7356-1779
+  - orcid:0000-0002-6601-2165
+creator_label:
+  - Nicolas Matentzoglu
+  - Chris Mungall
+
+author_id:
+  - orcid:0000-0002-7356-1779
+author_label:
+  - Nicolas Matentzoglu
+
+reviewer_id:
+  - orcid:0000-0001-9114-8737
+reviewer_label:
+  - Melissa Haendel
+
+# Source information
+subject_source: obo:mondo.owl
+subject_source_version: http://purl.obolibrary.org/obo/mondo/releases/2024-01-03/mondo.owl
+object_source: obo:ncit.owl
+object_source_version: http://purl.obolibrary.org/obo/ncit/releases/2024-01/ncit.owl
+
+# Mapping provenance
+mapping_tool: https://github.com/mapping-commons/sssom-py
+mapping_tool_version: "0.4.0"
+mapping_date: "2024-01-10"
+mapping_provider: https://monarchinitiative.org/
+
+# Sample mappings (demonstrating that mappings are preserved but not part of FAIR Mappings output)
+mappings:
+  - subject_id: MONDO:0005015
+    subject_label: diabetes mellitus
+    predicate_id: skos:exactMatch
+    object_id: NCIT:C2985
+    object_label: Diabetes Mellitus
+    mapping_justification: semapv:ManualMappingCuration
+    confidence: 0.95
+
+  - subject_id: MONDO:0004992
+    subject_label: cancer
+    predicate_id: skos:exactMatch
+    object_id: NCIT:C9305
+    object_label: Malignant Neoplasm
+    mapping_justification: semapv:ManualMappingCuration
+    confidence: 0.90
+
+  - subject_id: MONDO:0005301
+    subject_label: multiple sclerosis
+    predicate_id: skos:exactMatch
+    object_id: NCIT:C3243
+    object_label: Multiple Sclerosis
+    mapping_justification: semapv:LexicalMatching
+    confidence: 0.85
+    match_string:
+      - multiple sclerosis

tests/input/examples/fair_mappings_metadata/output/linkmlmap-fair-mappings.yaml

@@ -0,0 +1,41 @@
+id: https://example.org/transformations/gene-expression-mapping
+name: Gene Expression Data Transformation
+description: 'A transformation specification that maps gene expression data from
+
+  a legacy format to a standardized biomedical data model.
+
+
+  This specification includes class derivations for:
+
+  - GeneExpressionMixin (from GeneExpressionRecord)
+
+  - OrganismTaxon (from SampleMetadata)
+
+  - Study (from ExperimentInfo)
+
+
+  And enum derivations for:
+
+  - ExpressionLevelCategory (from LegacyExpressionLevel)'
+publication_date: '2024-01-15'
+license: https://creativecommons.org/licenses/by/4.0/
+version: 1.0.0
+mapping_method: semapv:ManualMappingCuration
+documentation: https://example.org/docs/gene-expression-mapping
+content_url: https://example.org/content/gene-expression-mapping.yaml
+creator:
+- id: orcid:0000-0002-1234-5678
+  name: Jane Doe
+  type: Person
+author:
+- id: orcid:0000-0003-9876-5432
+  name: John Smith
+  type: Person
+reviewer:
+- id: orcid:0000-0001-1111-2222
+  name: Alice Johnson
+  type: Person
+subject_source:
+  name: legacy_gene_expression.yaml
+object_source:
+  name: biolink_model.yaml

tests/input/examples/fair_mappings_metadata/output/sssom-fair-mappings.yaml

@@ -0,0 +1,12 @@
+id: http://purl.obolibrary.org/obo/mondo/mappings/mondo_exactmatch_ncit.sssom.tsv
+name: Mondo to NCIT Exact Mappings
+description: 'This mapping set contains exact mappings between Mondo Disease Ontology
+
+  and NCI Thesaurus (NCIT) terms. These mappings were curated by the
+
+  Monarch Initiative team to support disease data integration.'
+version: '2024-01-15'
+license: https://creativecommons.org/publicdomain/zero/1.0/
+publication_date: '2024-01-15'
+type: sssom
+mapping_method: https://github.com/mapping-commons/sssom-py