ORDO Mappings

- Delete: Unnecessary files and goals: temp analysis, getting of ordo.owl and querying it via SPARQL. - Add: SSSOM validation - Update: Replaced ad hoc SSSOM utilities with the ones from sssom-py
monarch-initiative · Apr 7, 2024 · c4aa5c0 · c4aa5c0
1 parent 12dc3d1
commit c4aa5c0
Show file tree

Hide file tree

Showing 8 changed files with 90 additions and 77 deletions.
diff --git a/a.txt b/a.txt
diff --git a/config/icd11.sssom-metadata.yml b/config/icd11.sssom-metadata.yml
@@ -1,12 +1,14 @@
-creator_id: 0000-0002-2906-7319
+creator_id: orcid:0000-0002-2906-7319
 curie_map:
   Orphanet: http://www.orpha.net/ORDO/Orphanet_
   icd11.foundation: http://id.who.int/icd/entity/
   oboInOwl: http://www.geneontology.org/formats/oboInOwl#
+  orcid: https://orcid.org/
   owl: http://www.w3.org/2002/07/owl#
   rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns#
   rdfs: http://www.w3.org/2000/01/rdf-schema#
   semapv: https://w3id.org/semapv/
   skos: http://www.w3.org/2004/02/skos/core#
   sssom: https://w3id.org/sssom/
 license: http://w3id.org/sssom/license/unspecified
+mapping_provider: https://www.orpha.net/
diff --git a/makefile b/makefile
@@ -1,12 +1,19 @@
 .DEFAULT_GOAL := all
-.PHONY: all release clean
+.PHONY: all release clean ontology mappings mappings-validate help
 TODAY ?=$(shell date +%Y-%m-%d)
 VERSION=v$(TODAY)
 SOURCE_URL=https://icd11files.blob.core.windows.net/tmp/whofic-2023-04-08.owl.gz
 
 
 # MAIN COMMANDS / GOALS ------------------------------------------------------------------------------------------------
-all: tmp/output/release/icd11foundation.owl tmp/output/release/ordo-icd11.sssom.tsv
+all: ontology mappings
+
+ontology: tmp/output/release/icd11foundation.owl
+
+mappings: mappings-validate
+
+mappings-validate: tmp/output/release/ordo-icd11.sssom.tsv
+	sssom validate $<
 
 clean:
 	rm -rf tmp/
@@ -42,13 +49,7 @@ tmp/input/Orphanet_Nomenclature_Pack_EN/ORPHA_ICD11_mapping_en_newversion_2023.x
 	wget https://www.orphadata.com/data/nomenclature/packs/Orphanet_Nomenclature_Pack_EN.zip -O tmp/input/Orphanet_Nomenclature_Pack_EN.zip
 	unzip tmp/input/Orphanet_Nomenclature_Pack_EN.zip -d tmp/input/Orphanet_Nomenclature_Pack_EN
 
-tmp/input/ordo.owl: | tmp/input/
-	wget http://www.orphadata.org/data/ORDO/ordo_orphanet.owl -O $@
-
-tmp/output/icd11mms-exact-matches.tsv: tmp/input/ordo.owl
-	robot query -i $< --query src/icd11mms-exact-matches.sparql $@
-
-tmp/output/release/ordo-icd11.sssom.tsv: tmp/input/Orphanet_Nomenclature_Pack_EN/ORPHA_ICD11_mapping_en_newversion_2023.xml tmp/output/icd11mms-exact-matches.tsv | tmp/output/release/
+tmp/output/release/ordo-icd11.sssom.tsv: tmp/input/Orphanet_Nomenclature_Pack_EN/ORPHA_ICD11_mapping_en_newversion_2023.xml | tmp/output/release/
 	python3 src/mappings.py \
 	--input-nomenclature-xml tmp/input/Orphanet_Nomenclature_Pack_EN/ORPHA_ICD11_mapping_en_newversion_2023.xml \
 	--input-sssom-config config/icd11.sssom-metadata.yml \

diff --git a/requirements-unlocked.txt b/requirements-unlocked.txt
@@ -1,2 +1,3 @@
 pandas
 pyyaml
+sssom
diff --git a/requirements.txt b/requirements.txt
@@ -1,14 +1,60 @@
-distlib==0.3.6
-filelock==3.9.0
-numpy==1.25.1
-pandas==2.0.3
-pbr==5.11.1
-platformdirs==3.1.0
-python-dateutil==2.8.2
-pytz==2023.3
+annotated-types==0.6.0
+attrs==23.2.0
+certifi==2024.2.2
+charset-normalizer==3.3.2
+click==8.1.7
+curies==0.7.9
+Deprecated==1.2.14
+deprecation==2.1.0
+distlib==0.3.8
+exceptiongroup==1.2.0
+filelock==3.13.1
+hbreader==0.9.1
+idna==3.6
+importlib_resources==6.4.0
+iniconfig==2.0.0
+isodate==0.6.1
+json-flattener==0.1.9
+jsonasobj2==1.0.4
+jsonschema==4.21.1
+jsonschema-specifications==2023.12.1
+linkml-runtime==1.7.5
+networkx==3.3
+numpy==1.26.4
+packaging==24.0
+pandas==2.2.1
+pansql==0.0.1
+pbr==6.0.0
+platformdirs==4.2.0
+pluggy==1.4.0
+prefixcommons==0.1.12
+prefixmaps==0.2.3
+pydantic==2.6.4
+pydantic_core==2.16.3
+pyparsing==3.1.2
+pytest==8.1.1
+pytest-logging==2015.11.4
+python-dateutil==2.9.0.post0
+PyTrie==0.4.0
+pytz==2024.1
 PyYAML==6.0.1
+rdflib==7.0.0
+referencing==0.34.0
+requests==2.31.0
+rpds-py==0.18.0
+scipy==1.13.0
 six==1.16.0
-stevedore==5.0.0
-tzdata==2023.3
-virtualenv==20.20.0
+sortedcontainers==2.4.0
+SPARQLWrapper==2.0.0
+SQLAlchemy==2.0.29
+sssom==0.4.6
+sssom-schema==0.15.2
+stevedore==5.1.0
+tomli==2.0.1
+typing_extensions==4.11.0
+tzdata==2024.1
+urllib3==2.2.1
+validators==0.28.0
+virtualenv==20.25.0
 virtualenv-clone==0.5.7
+wrapt==1.16.0
diff --git a/src/icd11mms-exact-matches.sparql b/src/icd11mms-exact-matches.sparql
diff --git a/src/mappings.py b/src/mappings.py
@@ -1,11 +1,19 @@
-"""Extract mappings"""
+"""Extract mappings
+
+todo's (minor):
+ 1. icd11.sssom-metadata.yml: Generalize it by removing the 'mapping_provider' field, which links to Orphanet, and add
+ that in dynamically just for these ORDO mappings.
+"""
 from argparse import ArgumentParser
 from pathlib import Path
 import xml.etree.ElementTree as eleTree
+from typing import Dict
 
+import curies
 import pandas as pd
-
-from utils import write_sssom
+import yaml
+from sssom import MappingSetDataFrame
+from sssom.writers import write_table
 
 SRC_DIR = Path(__file__).parent
 PROJECT_DIR = SRC_DIR.parent
@@ -131,7 +139,14 @@ def run(
 
     # Get only columns we care about
     df = df[['subject_id', 'subject_label', 'predicate_id', 'object_id']]
-    write_sssom(df, input_sssom_config, outpath)
+
+    # Write file
+    with open(input_sssom_config, 'r') as yaml_file:
+        metadata: Dict = yaml.load(yaml_file, Loader=yaml.FullLoader)
+    converter = curies.Converter.from_prefix_map(metadata['curie_map'])
+    msdf: MappingSetDataFrame = MappingSetDataFrame(converter=converter, df=df, metadata=metadata)
+    with open(outpath, 'w') as f:
+        write_table(msdf, f)
 
 
 def cli():

diff --git a/src/temp_compare_matches_owl_and_nomenclature.py b/src/temp_compare_matches_owl_and_nomenclature.py