Reverse engineering large-scale genetic networks: synthetic versus real data

Journal of genetics

Mei Xiao¹  Wu Zhang1¹  Luwen Zhang¹  Yong Wang² 

[1] School of Computer Engineering and Science, Shanghai University, 149 Yanchang Road, Zhabei District, Shanghai 200072, People’s Republic of China$$;Academy of Mathematics and Systems Science, Chinese Academy of Sciences (CAS), 52 San Lihe Road, Xicheng District 100864, Beijing, People’s Republic of China$$

关键词: gene regulatory network;    single gene perturbation;    linear model;    stepwise;    simulated network.;   

DOI  :  

学科分类：生物科学（综合）

来源: Indian Academy of Sciences

【 摘 要 】

Development of microarray technology has resulted in an exponential rise in gene expression data. Linear computational methods are of great assistance in identifying molecular interactions, and elucidating the functional properties of gene networks. It overcomes the weaknesses of in vivo experiments including high cost, large noise, and unrepeatable process. In this paper, we propose an easily applied system, Stepwise Network Inference (SWNI), which integrates deterministic linear model with statistical analysis, and has been tested effectively on both simulated experiments and real gene expression data sets. The study illustrates that connections of gene networks can be significantly detected via SWNI with high confidence, when single gene perturbation experiments are performed complying with the algorithm requirements. In particular, our algorithm shows efficiency and outperforms the existing ones presented in this paper when dealing with large-scale sparse networks without any prior knowledge.

【 授权许可】

Unknown