07. January 2016, 16:15 p.m.

Ernst-Abbe-Platz 2, seminar room 3423

NetGenerator: Inference of gene regulatory networks from time-series data

Dr. Sebastian Henkel
(BioControl Jena GmbH, Jena)

Inference, i.e. identification, of gene-regulatory networks (GRNs) is important for understanding behaviour and potential treatment of biological systems. Since GRNs are complex and dynamical, appropriate methods and algorithms are needed for constructing models describing these dynamics. Algorithms based on heuristic approaches may reduce the effort in parameter identification and computational time.

The NetGenerator algorithm, a heuristic for network inference, is presented and described in detail. It automatically generates a system of differential equations modelling structure and dynamics of the network based on time-resolved gene expression data. The algorithm is founded on the consideration of important biological properties: explicit stimuli (inputs), combination of hierarchy and feedback, sparsity and negligible interconversion. Also, structural prior knowledge can be provided to support the inference process.
After motivating important elements of NetGenerator, the application of the algorithm is demonstrated by an example system.

Finally, limitations of the algorithm are discussed and current challenges as well as potential future extensions are presented.