Topic: Contributed on High-dimensional latent variable models Title: Variable selection in binomial regression with latent Gaussian field models for analysis of epigenetic data Authors:  Aliaksandr Hubin - NMBU (Norway) [presenting]
Geir Olve Storvik - University of Oslo (Norway)
Abstract: Epigenetic observations data is represented by the total amount of reads from a particular cell and the amount of methylated reads, which are reasonable to model via a Binomial distribution. There are numerous factors that might influence the probability of success from a particular region. We might also expect spatial dependence of these probabilities. We incorporate dependence on the covariates and spatial dependence of probability of being methylated for observation from a particular cell by means of a binomial regression with latent Gaussian field model. We use INLA approach for calculating posterior marginal likelihoods for fixed models and carry out efficient MCMC with locally optimized mode jumping proposals across the models in order to draw from the posterior distributions of parameters and models jointly. During these MCMC walks we also simultaneously perform model selection with respect to different criteria to discover the optimal choice of covariates that influence methylation structure in the regions along the genome.