HIMA2: high-dimensional mediation analysis and its application in epigenome-wide DNA methylation data

• DOI: 10.1186/s12859-022-04748-1
• Publication Date: 2022-07-25T11:04:02Z
• Authors: Chamila Perera; Haixiang Zhang; Yinan Zheng; Lifang Hou; Annie Qu; Cheng Zheng; Ke Xie; Lei Liu
• Keywords: False Discovery Rate; Multiple comparisons problem

Abstract Mediation analysis plays a major role in identifying significant mediators the pathway between environmental exposures and health outcomes. With advanced data collection technology for large-scale studies, there has been growing research interest developing methodology high-dimensional mediation analysis. In this paper we present HIMA2, an extension of HIMA method (Zhang Bioinformatics 32:3150–3154, 2016). First, proposed HIMA2 reduces dimension to manageable level based on sure independence screening (SIS) (Fan J R Stat Soc Ser B 70:849–911, 2008). Second, de-biased Lasso procedure is implemented estimating regression parameters. Third, use multiple-testing accurately control false discovery rate (FDR) when testing hypotheses. We demonstrate its practical performance using Monte Carlo simulation studies apply our identify DNA methylation markers which mediate from smoking reduced lung function Coronary Artery Risk Development Young Adults (CARDIA) Study.