FAIR Service Descriptions: enriching life science SPARQL endpoints

Abstract

SPARQL service descriptions allow for rich information schemas describing the data inside SPARQL endpoints. Rewriting information schema (re)-discovery queries to queries using an existing one can give major performance benefits. Rich service descriptions have many use cases beyond query rewriting. A significant challenge for users of SPARQL endpoints is discovering the shape and quantity of the data exposed inside them.

-

Using the Service Description [ 1 ], VoID [ 2 ] and VoID-Ext [ 3 ] vocabularies. We store these in in-dependant named graphs, which we always name as address of the SPARQL endpoint + (N. Redaschi) is. This is because there are more than 140 billion distinct triples in UniProt. Of course having such a SD is not enough as the people who are used to using such queries won’t change to use a diferent query on an ”information schema” by default. This means we need to rewrite the query (listing:1) to a query in the form of (listing:3). Query rewriting needs to take into account variations in prefix, white-space and variable naming. We solve this by using a SPARQL parser from the RDF4j project use the abstract SPARQL algebra for the query matching and rewrite. The original query with is redirected to a new location with a new query (http 301).

Listing 1: ”Count distinct classes used in a SPARQL endpoint.”

SELECT ( COUNT ( DISTINCT ? c l a s s ) AS ? c l a s s e s ) WHERE { ? s u b j e c t a ? c l a s s . }

Listing 2: ”Simple pipeline to count the unique classes in an ntriples file.”

s o r t −u a l l _ t r i p l e s _ i n _ u n i p r o t . n t | g r e p r d f : type | s o r t −u | wc − l Listing 3: ”Rewritten SPARQL query to retrieve the count of the distinct classes in the endpoint.” SELECT ( COUNT ( DISTINCT ? c l a s s e s R a w ) AS ? c l a s s e s ) FROM < h t t p : / / s p a r q l . u n i p r o t . o r g / . w e l l −known / v o i d > WHERE { [ ] < h t t p : / / r d f s . o r g / n s / v o i d # c l a s s > ? c l a s s e s R a w . } Acknowledgments The Swiss-Prot group is part of the SIB Swiss Institute of Bioinformatics and of the UniProt Consortium. Swiss-Prot group activities are supported by the Swiss Federal Government through the State Secretariat for Education, Research and Innovation SERI and UniProt is supported by the National Eye Institute (NEI), National Human Genome Research Institute (NHGRI), National Heart, Lung, and Blood Institute (NHLBI), National Institute on Aging (NIA), National Institute of Allergy and Infectious Diseases (NIAID), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institute of General Medical Sciences (NIGMS), National Institute of Mental Health (NIMH), and National Cancer Institute (NCI) of the National Institutes of Health (NIH) under grant U24HG007822.

References

[1] Sparql 1 .1 service description, 2013 . URL: https://www.w3.org/TR/ sparql11-service-description/.

[2] M. H. J. Z. Keith Alexander , Richard Cyganiak, Describing linked datasets with the void vocabulary , 2011 . URL: https://www.w3.org/TR/void/.

[3] E. Mäkelä , Aether - generating and viewing extended void statistical descriptions of rdf datasets , in: V. Presutti , E. Blomqvist, R. Troncy , H. Sack , I. Papadakis , A . Tordai (Eds.), The Semantic Web: ESWC 2014 Satellite Events , Springer International Publishing, Cham, 2014 , pp. 429 - 433 .