Overview

Following the advent of new high throughput technologies including gene expression microarrays, protein arrays, and more recently high throughput sequencing, statistical genomics has emerged as an important research area within the field of statistics. A major challenge is to formulate and implement solutions to questions stemming from the collection and analysis of data arising in experimental molecular biology. These questions stem, in the narrow sense, from the analysis of individual components of biological systems (e.g., structural and functional genomics, proteomics, metabolomics), the analysis of the interaction between components (biological networks) or even at the level of entire biological systems. It is apparent that detailed statistical methods are needed to integrate and carry out inferences for such data, and that many of the methods require advanced computation, such as Markov Chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC). For MCMC/SMC many theoretical issues are largely settled, but practical issues remain, and increase in difficulty with model complexity. Typically experimental biological data are incomplete, high-dimensional and subject to considerable observation error, and thus tailored computational approaches are required; this therefore represents a challenging arena for MCMC practitioners.

This five-day workshop will provide a networking opportunity to researchers and students in statistical genomics. Discussions and a poster session will stimulate exchanges between participants.