Comparison of gene co-expression networks and bayesian networks

Authors:
Saurabh Nagrecha;Pawan J. Lingras;Nitesh V. Chawla
Affiliations:
Department of Computer Science and Engineering, University of Notre Dame, Indiana;Department of Mathematics and Computing Science, Saint Mary's University, Halifax, Nova Scotia, Canada;Department of Computer Science and Engineering, University of Notre Dame, Indiana
Venue:
ACIIDS'13 Proceedings of the 5th Asian conference on Intelligent Information and Database Systems - Volume Part I
Year:
2013

Citing 1
Cited 0

SOM-based class discovery exploring the ICA-reduced features of microarray expression profiles: Research Paper

Comparative and Functional Genomics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Inferring genetic networks is of great importance in unlocking gene behaviour, which in turn provides solutions for drug testing, disease resistance, and many other applications. Dynamic network models provide room for handling noisy or missing prelearned data. This paper discusses how Dynamic Bayesian Networks compare against coexpression networks as discussed by Zhang and Horvath [1]. These shall be tested out on the genes of yeast Saccharomyces cerevisiae. A method is then proposed to get the best out of the strengths of both models, namely, the causality inference from Bayesian networks and the scoring method from a modified version of Zhang and Horvath's method.