A two-stage methodology for gene regulatory network extraction from time-course gene expression data

Authors:
Zeke S. H. Chan;N. Kasabov;Lesley Collins
Affiliations:
Knowledge Engineering and Discovery Research Institute (KEDRI), Auckland University of Technology, Auckland, New Zealand;Knowledge Engineering and Discovery Research Institute (KEDRI), Auckland University of Technology, Auckland, New Zealand;Allan Wilson Centre for Molecular Ecology and Evolution, Massey University, Palmerston North, New Zealand
Venue:
Expert Systems with Applications: An International Journal
Year:
2006

Citing 1
Cited 3

Using genetic algorithm to support portfolio optimization for index fund management

Expert Systems with Applications: An International Journal

Cooperativeness prediction in P2P networks

Expert Systems with Applications: An International Journal
Personalised modelling for multiple time-series data prediction: a preliminary investigation in Asia Pacific stock market indexes movement

ICONIP'08 Proceedings of the 15th international conference on Advances in neuro-information processing - Volume Part I
Genetic Networks and Soft Computing

IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)

Quantified Score

Hi-index	12.05

Visualization

Abstract

The discovery of gene regulatory networks (GRN) from time-course gene expression data (gene trajectory data) is useful for (1) identifying important genes in relation to a disease or a biological function; (2) gaining an understanding on the dynamic interaction between genes; (3) predicting gene expression values at future time points and accordingly; (4) predicting drug effect over time. In this paper, we propose a two-stage methodology that is implemented in the software 'Gene Network Explorer (GNetXP)' for extracting GRNs from gene trajectory data. In the first stage, we apply a hybrid Genetic Algorithm and Expectation Maximization algorithm on clustering the large number of gene trajectories using the mixture of multiple linear regression models for fitting the trajectory data. In the second stage, we apply the Kalman Filter to identify a set of first-order differential equations that describe the dynamics of the representative trajectories, and use these equations for discovering important gene interactions and predicting gene expression values at future time points. The proposed method is demonstrated on the human fibroblast response gene expression data.