Section: Application Domains

Biomedical Knowledge Discovery

Participants : Marie-Dominique Devignes [contact person] , David Ritchie.

This project is in collaboration with the Orpailleur Team.

Increasing amounts of biomedical data provided as Linked Open Data (LOD) offer novel opportunities for knowledge discovery in biomedicine. We published an approach for selecting, integrating, and mining LOD with the goal of discovering genes responsible for a disease [46] . The selection step relies on a set of choices made by a domain expert to isolate relevant pieces of LOD. Because these pieces are potentially not linked, an integration step is required to connect unlinked pieces. The resulting graph is subsequently mined using Inductive Logic Programming (ILP) that presents two main advantages. First, the input format compliant with ILP (first order logic) is close to the format of LOD (RDF triples). Second, domain knowledge can be added to this input and used during the induction step. We have applied this approach to the characterization of genes responsible for intellectual disability. For this real-world use case, we could evaluate ILP results and assess the contribution of domain knowledge. Our ongoing efforts explore how the combination of rules coming from distinct theories can improve the prediction accuracy [45] , [55] .