bgx logo

Bayesian Integrative Genomics

Phase I: Funded by the BBSRC Exploiting Genomics initiative, from May 2002 to February 2007.

Phase II: A collaborative programme started in 2008, coordinated by Prof Sylvia Richardson.

ESS++: Bayesian variable selection for linear regression using evolutionary Monte Carlo

ESS++ is a C++ implementation of a fully Bayesian variable selection approach for single and multiple response linear regression. ESS++ works well both when the number of observations is larger than the number of predictors and in the 'large p, small n' case. In the current version (0.1), ESS++ can handle several hundred observations, thousands of predictors and a few responses simultaneously. The core engine of ESS++ for the selection of relevant predictors is based on evolutionary Monte Carlo. The C++ implementation of ESS++ is open source and distributed according to the GNU GPL licence.

Release 0.1

A zipped tar archive containing software and documentation is available by clicking this link. Full instructions for installation, compilation, running and testing the code are provided in the documentaton, which can also be downloaded separately here. For full details of the underlying algorithm, please see the papers referenced within the documentation.

Important Release Note:

ESS is now part of GUESS package. To get the latest version of the code and the documentation, please visit GUESS webpage.