Hybrid Approach to Inferring Gene Networks with Expression Profile and Gene Ontology

L. Jing, Y. Liu, and P. Bachman (USA)


Hybrid approach, Gene regulation network, Expression profile, Gene Ontology.


Gene regulatory networks have long been studied in model organisms as a means of identifying functional relation ships among genes or their corresponding gene products. Despite many existing methods for genome-wide inferring of such networks, solutions to the gene regulatory net works problem are however not trivial. Here we present HAEO, a hybrid approach with gene expression profiles and Gene Ontology. HAEO firstly identifies the functional gene groups by overlapping clustering the genes based on their annotated terms in Gene Ontology, esp., molecu lar functions and biological processes. Next, HAEO uses Bayesian networks method to infer local gene regulatory networks (which are more consistent and can be treated as the local components of the global gene regulatory net works) for each group based on gene expression profiles. Finally, these local regulatory networks are naturally linked by the overlap genes between groups to compose a global network. HAEO shows promise in effectively and effi ciently inferring gene regulatory networks on Yeast Sac charomyces cerevisiae cell-cycle microarray data and Gene Ontology.

Important Links:

Go Back