Inferring Topology and Dynamical Properties of Genome-wide Regulatory Networks
Dr Richard Bonneau ( Center for Genomics and Systems Biology, New York University )
-
14:00 22nd May 2009 ( week Week 4, Trinity Term 2009 )Room 478, Computing Laboratory
Inferelator-1 is a network reconstruction algorithm that uses an expression dataset to infer genome wide regulatory networks.
Specifically, it learns a system of ordinary differential equations that describe the rate of change in transcription, of
each gene, as a function of relevant predictors.
Here, we 1) implement a resampling technique, Inferelator-1.1, to produce an ensemble of networks that encapsulate potentially different network topologies and kinetic parameters; and 2) refine this ensemble, via a MCMC dynamical modeling method, Inferelator-2, to learn high likelihood RNs. Initial results indicate significant improvement in our ability to correctly learn topology, and to model dynamics.
[1] Bonneau R, et.al.(2006) The Inferelator: an algorithm for learning parsimonious regulator networks from systems biology data sets de novo. Genome Biology 7, R36
[2] Bonneau R, et. al. (2007) A Predictive Model for Transcriptional Control of Physiology in a Free Living Cell. Cell 131, 7:1354-1365
Here, we 1) implement a resampling technique, Inferelator-1.1, to produce an ensemble of networks that encapsulate potentially different network topologies and kinetic parameters; and 2) refine this ensemble, via a MCMC dynamical modeling method, Inferelator-2, to learn high likelihood RNs. Initial results indicate significant improvement in our ability to correctly learn topology, and to model dynamics.
[1] Bonneau R, et.al.(2006) The Inferelator: an algorithm for learning parsimonious regulator networks from systems biology data sets de novo. Genome Biology 7, R36
[2] Bonneau R, et. al. (2007) A Predictive Model for Transcriptional Control of Physiology in a Free Living Cell. Cell 131, 7:1354-1365