Avoiding spurious feedback loops in the reconstruction of gene regulatory networks with dynamic Bayesian networks

Publication Name
Pattern Recognition in Bioinformatics
Publisher
Springer Verlag
ISBN
978-3-642-04030-6
Abstract
Feedback loops and recurrent structures are essential to the regulation and stable control of complex biological systems. The application of dynamic as opposed to static Bayesian networks is promising in that, in principle, these feedback loops can be learned. However, we show that the widely applied BGe score is susceptible to learning spurious feedback loops, which are a consequence of nonlinear regulation and autocorrelation in the data. We propose a non-linear generalisation of the BGe, based on a mixture model, and demonstrate that this approach successfully represses spurious feedback loops.
Year
2009
Category
Book Chapter