Differential network analysis via lasso penalized D-trace loss

• 成果类型: Article
• 署名作者: Yuan, Huili; Xi, Ruibin; Chen, Chong; Deng, Minghua
• 署名单位: Peking University
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asx049
• 发表日期: 2017
• 页码: 755770
• 关键词: Covariance Estimation; regulatory networks; selection; MODEL; Consistency; mutations; cancer; kegg

摘要

Biological networks often change under different environmental and genetic conditions. In this paper, we model network change as the difference of two precision matrices and propose a novel loss function called the D-trace loss, which allows us to directly estimate the precision matrix difference without attempting to estimate the precision matrices themselves. Under a new irrepresentability condition, we show that the D-trace loss function with the lasso penalty can yield consistent estimators in high-dimensional settings if the difference network is sparse. A very efficient algorithm is developed based on the alternating direction method of multipliers to minimize the penalized loss function. Simulation studies and a real-data analysis show that the proposed method outperforms other methods.