Public health Genomics to Practice in Cardiovasculaur diseases

Researchers: Aki Vehtari, Pekka Marttinen, Tomi Peltola, Heikki Peura

The goal is to develop statistical data analysis methods for health genomics, epidemiology and health economics related to cardiovascular diseases (CVD). Even when focusing on one specific aspect of population or individual health, other competing risk factors and underlying determinants make the overall phenomena complex. Generally complex phenomena may have complex structure or apparently simple structure with a large number of elements (e.g. competing risk factors and large number of genes) producing emergent complexity. Modeling is challenging if known structural information is incomplete or if there is a large number of unknown parameters compared to the amount of observations available. Ever increasing computing performance makes it possible to use more complex models to model phenomena which are inherently complex containing nonlinearities, interactions and network like structures. Bayesian approach provides consistent and flexible way to combine available structural information and uncertain observations. Methodological research made will support the research made in the other sub-projects, including more careful dissection of genome wide association data, and modeling the genetic structure of the Finnish population and its association with spatial variation in CVD morbidity. Advanced Bayesian prediction and latent variable methods will be used to model environmental parameters and combine genetical and spatiotemporal models. Use of complex models causes need for research, for example, in estimation of relevance of potential explanatory covariates, estimation of predictive performance, and visualisation and interpretation of complex interacting effects.

The research is part of Computational Health.

The project is part of Finnish Academy funded Public health genomics to practice in cardiovascular diseases consortium lead by Docent Markus Perola, Institute of Molecular Medicine, Finland (FIMM). Partners in the consortium are

• Institute of Molecular Medicine, Finland (FIMM)
• Sanger Institute, Hinxton, UK
• National Institute for Health and Welfare
• McGill University, Montreal, Canada
• University of Helsinki
• Dept of Biomedical Engineering and Computational Science (BECS), Aalto University