Differential regulation enrichment analysis via the integration of transcriptional regulatory network and gene expression data

Motivation: Although many gene set analysis methods have been proposed to explore associations between a phenotype and a group of genes sharing common biological functions or involved in the same biological process, the underlying biological mechanisms of identified gene sets are typically unexplained.Results: We propose a method called Differential Regulation-based enrichment Analysis for GENe sets (DRAGEN) to identify gene sets in which a significant proportion of genes have their transcriptional regulatory patterns changed in a perturbed phenotype. We conduct comprehensive simulation studies to demonstrate the capability of our method in identifying differentially regulated gene sets. We further apply our method to three human microarray expression datasets, two with hormone treated and control samples and one concerning different cell cycle phases. Results indicate that the capability of DRAGEN in identifying phenotype-associated gene sets is significantly superior to those of four existing methods for analyzing differentially expressed gene sets. We conclude that the proposed differential regulation enrichment analysis method, though exploratory in nature, complements the existing gene set analysis methods and provides a promising new direction for the interpretation of gene expression data.Availability and implementation: The program of DRAGEN is freely available at http://bioinfo.au.tsinghua.edu.cn/dragen/.Contact: ruijiang@tsinghua.edu.cn or jiang@cs.ucr.eduSupplementary information: Supplementary data are available at Bioinformatics online.

动机：尽管已经提出了许多基因集分析方法来探索表型与一组具有共同生物学功能或参与同一生物学过程的基因之间的关联，但所识别基因集的潜在生物学机制通常未得到解释。 结果：我们提出了一种名为基于差异调控的基因集富集分析（DRAGEN）的方法，用于识别在受扰表型中很大比例的基因其转录调控模式发生改变的基因集。我们进行了全面的模拟研究，以证明我们的方法在识别差异调控基因集方面的能力。我们进一步将我们的方法应用于三个人类微阵列表达数据集，其中两个包含激素处理和对照样本，一个涉及不同细胞周期阶段。结果表明，DRAGEN在识别与表型相关的基因集方面的能力明显优于四种现有的分析差异表达基因集的方法。我们得出结论，所提出的差异调控富集分析方法虽然具有探索性，但对现有的基因集分析方法是一种补充，并为基因表达数据的解释提供了一个有前景的新方向。 可用性和实现：DRAGEN程序可在http://bioinfo.au.tsinghua.edu.cn/dragen/免费获取。 联系人：ruijiang@tsinghua.edu.cn或jiang@cs.ucr.edu 补充信息：补充数据可在Bioinformatics在线获取。