This study includes two projects. The aim of project 1 is to develop machine learning (ML) models that integrate multi-omics data such as genetic, epigenetic, RNA, blood and inflammation markers, alongside lifestyle and patient information to refine risk prediction, prognosis and diagnosis of specific chronic age associated diseases. The project will study a) how well such ML models work, b) which age associated diseases can be predicted by such models, c) what the value is of polygenic scores in such models, d) what the value is of methylation data in such models, and e) what the value is of increasing omics layers in such models.
In project 2, the aim is to train algorithms (re-using models developed in project 1) on multiple layers of omics biomarkers, alongside clinical information, in order to develop a measure of patient vulnerability/general frailty syndrome.