Dr. Thierry Chekouo, Assistant Professor at the University of Calgary, will present:

“Integrative Bayesian Methods for Omics and Imaging Data: Application to Cardiovascular and Glioblastoma Diseases”

Abstract: The problem of integrating multi-omics (including radiomics) data from multiple sources is an important one in modern biomedical research. Multi-omics data generated for the same set of samples can help in unraveling the mechanisms underlying the biological condition of interest, and enhance our understanding of the pathobiology of complex diseases. In this talk, I will mainly focus on two projects.

i) First project. Motivated by the need for exploring other risk factors of atherosclerotic cardiovascular disease (ASCVD), we propose a Bayesian integrative model (BIPnet) for integrative analysis of clinical, demographic and omics data to identify genetic variants, genes, and gene networks likely contributing to 10-year ASCVD risk in healthy adults. We consider a factor analysis approach, and we identify simultaneously active components and important features within components using two nested layers of binary latent indicators. Our prior distributions can be extended to incorporate external grouping information.

ii) Second project. For the purpose of identifying pathways and corresponding genes that can explain the heterogeneity of the composition of the glioblastoma multiform (GBM) brain tumor, we propose a Bayesian hierarchical model for variable selection with a group structure in the context of correlated multivariate compositional response variables. We model the proportions of the tumor components within the tumor using a Dirichlet model by allowing for straightforward incorporation of available high-dimensional covariate information within a log-linear regression framework. We impose prior distributions that account for the overlapping structure between groups of covariates.

Simulations and applications to ASCVD and GBM data sets show the importance of our approaches.

All are Welcome.