maastrichtu-ids / food-claims-kg

🧅 RDF knowledge graph for food health claims

A RDF knowledge graph for food health claims.

The easiest way to explore the knowledge graph is to use the various API calls at http://grlc.io/api-git/MaastrichtU-IDS/food-claims-kg

The SPARQL endpoint is accessible at https://graphdb.dumontierlab.com/repositories/FoodHealthClaimsKG

Check the SPARQL queries

Queries from the grlc API can be checked and changed in the .rq files.

Variables starting with an underscore, e.g. _claimlabel, are asked to the user as parameter of the generated API call.

Run conversion scripts

Scripts run automatically with GitHub Actions at every push to the master branch.

Clone the repository:

git clone https://github.com/MaastrichtU-IDS/food-claims-kg.git
cd food-claims-kg

Install requirements:

pip3 install -r requirements.txt

Download the data:

Link to Google Sheet

mkdir -p data/output
wget -O data/food-claims-kg.xlsx "https://docs.google.com/spreadsheets/d/1RWZ6AlGB8m7PO5kjsbbbeI4ETLwvKLOvkrzOpl8zAM8/export?format=xlsx&id=1RWZ6AlGB8m7PO5kjsbbbeI4ETLwvKLOvkrzOpl8zAM8"

Define API key to query UMLS:

export UMLS_APIKEY=00000000000000

Run the conversion script:

python3 src/convert_to_rdf.py

Create metadata (HCLS statistics)

d2s metadata analyze https://graphdb.dumontierlab.com/repositories/FoodHealthClaimsKG -o metadata.ttl 

Examples of grlc API

API powered by grlc.io 🧅

IRI variable with defaults value (not working for our repo):

• https://github.com/albertmeronyo/lodapi/blob/master/dbpedia_test.rq
• API: http://grlc.io/api-git/albertmeronyo/lodapi#/dbpedia/get_dbpedia_test

String variable:

• https://github.com/CLARIAH/wp4-queries-hisco/blob/master/get-hisco-hiscam.rq
• API: http://grlc.io/api-git/CLARIAH/wp4-queries-hisco#/hisco/get_get_hisco_hiscam

Enumerate:

• https://github.com/CLARIAH/grlc-queries/blob/master/enumerate.rq
• API: http://grlc.io/api-git/CLARIAH/grlc-queries/#/default/get_enumerate

food-claims-kg's Issues

Reference with no link

Remove "http://doi.org/" references and use a prefix for ISBN identifiers.

Integrate FooDB

https://foodb.ca/downloads

Serve the SPARQL endpoint with an API

Use grlc API to define API calls converted as SPARQL queries with variables? https://github.com/CLARIAH/grlc

Or use d2s-api or RdfServices: https://github.com/amalic/rdfservices

Integrate ontologies from BioPortal

• UMLS, MEDDRA, GO, HPO - phenotypes (priority)

We just want mappings from UMLS to GO, MEDDRA and HPO

• ChemBL - food
• FOODON - food items / categories
• Mesh. - age
• SNOMED-CT - population/social condition
• MEDDRA - population/social condition