Estimating the regulatory potential of DNA methylation in Alzheimer's disease. (December 2021)

• Record Type: Journal Article
• Authors: Wang, Lily; Silva, Tiago C; Young, Juan; Martin, Eden R; Kunkle, Brian W.; Chen, Xi
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 3
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.049365
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Epigenome‐wide association studies (EWAS) often detect a large number of differentially methylated CpGs, many are located far from genes, complicating the interpretation of their functionalities. Therefore, there is a critical need to better understand the functional impact of these CpGs. Recent studies demonstrated methylated CpGs can affect gene transcription by either increase or decrease transcription factor binding strengths. To prioritize significant CpGs from EWAS, an integrative analysis that assesses the impact of CpG methylation on TF regulatory activities is proposed. Method: We developed a new method and software, MethReg, that analyzes matched DNA‐methylation and gene‐expression data, along with external transcription factor (TF) binding information, to evaluate, prioritize, and annotate CpG sites with high regulatory potential. By simultaneous modeling three key elements that contribute to gene transcription (CpG methylation, target gene expression and TF activity), MethReg identifies TF‐target gene associations that are present only in a subset of samples with high (or low) methylation levels at the CpG that influences TF activities, which can be missed in analyses that use all samples. Result: We performed a MethReg analysis for the ROSMAP Alzheimer's Disease (AD) dataset with DNA methylation and RNA‐seq data for 529 independent subjects. MethReg identified 60 methylation sensitive transcription factors, many of which are well‐known regulators for AD such as TCF12, SPI1, NR3C1, CEBPB, GABPA, and others. Conclusion: Although many of these significant TFs have been previously implicated in AD pathology, their specific roles in transcription regulation and the identification of their targets in AD remain to be investigated. Currently available tools only identify the TFs but do not consider CpGs or provide detailed information on the relevant target genes. In contrast, MethReg fills this critical gap by nominating plausible TF‐target associations that are mediated by DNA methylation. Therefore, MethReg analysis, which leverages additional gene expression data and provides more comprehensive information on transcription regulation for the TFs, complements existing approaches.