Abstract
We describe a method for the integration of high-throughput data from different sources. More specifically, iBATCGH is a package for the integrative analysis of transcriptomic and genomic data, based on a hierarchical Bayesian model. Through the specification of a measurement error model we relate the gene expression levels to latent copy number states which, in turn, are related to the observed surrogate CGH measurement via a hidden Markov model. Selection of relevant associations is performed employing variable selection priors that explicitly incorporate dependence information across adjacent copy number states. Posterior inference is carried out through Markov chain Monte Carlo techniques that efficiently explores the space of all possible associations. In this Chapter we review the model and present the functions provided in iBATCGH, an R package based on a C implementation of the inferential algorithm. Lastly, we illustrate the method via a case study on ovarian cancer
Original language | English |
---|---|
Title of host publication | Statistical Analysis for High-Dimensional Data |
Subtitle of host publication | The Abel Symposium 2014 |
Editors | Frigessi Arnoldo, Peter Buhlmann, Ingrid K. Glad, Mette Langaas, Sylvia Richardson, Marina Vannucci |
Publisher | Springer Verlag |
Pages | 105-123 |
ISBN (Electronic) | 978-3-319-27099-9 |
ISBN (Print) | 978-3-319-27097-5 |
DOIs | |
Publication status | Published - 1 Feb 2016 |