Work Package 7: Data Integration

Led by Pietro Spitali (LUMC)

With the main objective of providing FAIR data for analytics, during this year WP7 focused on: (1) the FAIRification of experimental data from different WP data partners, (2) the integration of these multimodal experimental data, and on the integration of these data with the knowledge map, and (3) we also started computational analyses based on artificial intelligence to answer questions on the molecular mechanisms underlying the genotype to brain phenotype correlation and get new hypotheses on the potential role of dystrophin.  

FAIRification of experimental data: FAIRification is occurring in two parallalel levels: (a) at DATASET level; and (b) at DATA level. At dataset level, we are implementing WPs’ metadata in the DDF FAIR Data Point. At data level, we are modelling BIND data re-using the European Joint Programme on Rare Diseases core semantic model used in rare disease patient registries and transforming these data to RDF knowledge graphs through the BIND models and developed custom scripts. 

Data and knowledge map integration: We are normalising WPs and knowledge map data models for knowledge and data integration as a single RDF knowledge graph. 

Analytics: We are doing preliminary work using sophisticated graph machine learning methods to generate prediction models for WP7 questions, and on proteomics-genetics expression analysis.