Within the context of Systems Biology there is a growing requirement for the development of mechanistic models to support reasoning about the structures of biochemical pathways. In addition to defining such models, a means of objectively assessing the validity of competing hypotheses regarding pathway structures based on experimental data and prior knowledge is essential. As the models themselves will have been identified from experimental observations for which there is significant variability it is advisable to adopt a consistent grammar for scientific reasoning that will take account of this uncertainty.

The Bayesian perspective is highly appropriate to enable consistent reasoning over mechanistic models of biological systems. However, given the intractable nature of the integrals required for an analytical Bayesian solution we are required to turn to Markov chain Monte Carlo or other approximating techniques to perform system identification and model-based reasoning. Practical solutions to these problems are critical to the utility and realism of systems biology models.

The goal of this workshop explores the main methodological and technical issues associated with performing Bayesian inference over mechanistic biochemical pathway models. It brings together experts in systems biology, statistical inference and machine learning (through the PASCAL networks).

For more information visit the Workshop website.