His main interest lies in developing statistical methods that connect large scale data sets (genomics, transcriptomics, metabolomics and clinically important phenotypes) in order to reveal novel molecular mechanisms underlying disease and infer causal effects of various risk factors on complex traits.