CaliBayes
Integration of GRID-based post-genomic data resources through Bayesian calibration of biological simulators
Welcome to the CaliBayes home page!
The CaliBayes project is concerned with the problem of estimating the kinetic parameters of dynamic biological simulation models (either deterministic or stochastic) encoded in SBML by using time course experimental data.
Please note that the CaliBayes webservices hosted at Newcastle are not fully functional currently, due to hardware problems. It is unlikely that this will be resolved satisfactorily since the funded project is over, however the architecture may be replicated by other users on their own hardware. The VirtualBox is still available as a demonstration of the system and available APIs and libraries.
The main aim of the CaliBayes project is to provide a powerful new tool for the academic community based on advanced Bayesian statistical inference technology, which enables inferences to be made about kinetic parameters within large and complex deterministic and stochastic network models of biochemical pathways and cell signalling systems. CaliBayes capitalises on the development of GRID-based modelling and simulation resources such as BASIS by building a higher-level computational web-services infrastructure using a service-oriented architecture (SOA) designed for integration of multiple post-genomic data resources. The project is a collaborative, multidisciplinary project involving the School of Mathematics & Statistics and the Institute of Ageing and Health at Newcastle University. CaliBayes was funded by the UK Biotechnology and Biological Sciences Research Council (BBSRC) under the Bioinformatics and e-Science Program (BBSB16550). The direct funding of the project ended December 2008, but the project continues as a collaboration between the Statistical Bioinformatics and Stochastic Systems Biology Research Group and the Centre for Integrated Systems Biology of Ageing and Nutrition (CISBAN). CaliBayes, together with its sister project BASIS, form the main computational infrastructure for modelling, simulation and inference within CISBAN.
The main outputs of the CaliBayes project are:
The CaliBayes software infrastructure, and associated Documentation
The public CaliBayes web-services - Please note that these web-services are not functioning currently due to hardware problems.
Model calibration tutorial - an introduction to the use of the public CaliBayes web services
New statistical technology for Bayesian model calibration and stochastic model emulation
A number of publications, talks and posters describing the research and our findings