Gene network inference using a swarm intelligence framework

Kyriakos Kentzoglanakis, Matthew Poole

Research output: Contribution to conferencePaperpeer-review

152 Downloads (Pure)


In this paper, we present a framework for inferring gene regulatory networks from gene expression time series. A model-based approach is adopted, according to which the quality of a candidate architecture is evaluated by assessing the ability of the corresponding trained model to reproduce the available dynamics. Candidate architectures are generated in the context of the ant colony optimization (ACO) meta-heuristic and model training is performed using particle swarm optimization (PSO). We propose a novel solution construction heuristic for artificial ants, based on growth and preferential attachment, in order to generate candidate structures that adhere to well-known gene network properties. Preliminary results using an artificial network demonstrate the potential of the framework to infer the underlying network architecture to a promising degree of success.
Original languageEnglish
Publication statusPublished - 2009
Event11th Annual Conference on Genetic and Evolutionary Computation - Montreal, Canada
Duration: 8 Jul 200912 Jul 2009


Conference11th Annual Conference on Genetic and Evolutionary Computation
Abbreviated titleGECCO '09


Dive into the research topics of 'Gene network inference using a swarm intelligence framework'. Together they form a unique fingerprint.

Cite this