Regulatory network inference: learning the structure of biological circuits from data:

We have developed a methodology for deriving transcriptional regulatory interactions on a genome-wide scale, and have applied the method to predict a large portion of the gene regulatory network of the archaea, Halobacterium NRC-1. The learned network is predictive, learned entirely from data de novo, and was used to successfully predict the global expression of Halobacterium under novel perturbations (not part of the original training set) with predictive power similar to that seen over the training set. Methodological advancements over earlier work include an explicit treatment of time such that the network model can be fit using both steady-state measurements and heterogeneous time series simultaneously. The method contains a novel means for learning binary logic interactions between regulators that requires no discretization of data. This work was done in tight collaboration with Nitin Baliga, Vestienn Thorsson , David Reiss and Lee Hood at the ISB. There are many interesting future directions on this project including: adding tighter control via user defined constraints and addition of known interactions to the current framework, use of proteomic and metabolomic data, using the networks for engineering and applying the method to subsequently more and more complex organisms. This program, the Inferelator, is a companion program to cMonkey.