2025/06/04 by Henriette Harmse, Harmse, Henriette, Haider Iqbal +5
Biochemistry, Genetics and Molecular Biology · Computer Science · Decision Sciences · #Biomedical Text Mining and Ontologies #Databases (cs.DB) #FOS: Computer and information sciences #Scientific Computing and Data Management #Semantic Web and Ontologies
paper · pdf · doi:10.48550/arxiv.2506.04286
openalex publication_date 2025/06/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
EMBL-EBI created OxO to enable users to map between datasets that are annotated with different ontologies. Mappings identified by the first version of OxO were not necessarily logically sound, lacked important provenance information such as author and reviewer, and could timeout or crash for certain requests. In this paper we introduce OxO2 to address these concerns. Provenance is addressed by implementing SSSOM, a mapping standard that defines provenance for mappings. SSSOM defines the conditions under which logical sound mappings can be derived and is implemented in OxO2 using Nemo, a Datalog rule engine. To ensure reasoning is performant and memory efficient, Nemo implements a number of strategies that ensures OxO2 will be stable for all requests. Due to these changes, OxO2 users will be able to integrate between disparate datasets with greater confidence.