Online Program Home
My Program

Abstract Details

Activity Number: 81 - New Development in Epigenome-Wide Association Studies
Type: Contributed
Date/Time: Sunday, July 29, 2018 : 4:00 PM to 5:50 PM
Sponsor: Section on Statistics in Genomics and Genetics
Abstract #329387
Title: De Novo Detection and Accurate Inference of Differentially Methylated Regions
Author(s): Keegan Korthauer* and Sutirtha Chakraborty and Yuval Benjamini and Rafael Irizarry
Companies: Dana-Farber Cancer Institute and Novartis and Hebrew University of Jerusalem and Harvard University
Keywords: differential methylation; epigenetics; bisulfite sequencing; generalized least squares; correlated error; variance-stabilizing transformation
Abstract:

A fundamental task in the analysis of methylation sequencing data is to detect Differentially Methylated Regions (DMRs), a key step in untangling the complex role of epigenetic modification in gene regulation. However, current computational approaches for detecting such differential regions do not provide accurate statistical inference. A major challenge in reporting uncertainty is that a genome-wide scan is involved in detecting regions, which needs to be properly accounted for. A further challenge is that sample sizes are limited due to the cost of the technology. We propose an approach that detects DMRs and assesses their uncertainty in a rigorous manner. Transformed methylation levels are modeled using generalized least squares while accounting for inter-individual and inter-loci variability. Significance of DMRs is assessed against a pooled null distribution that can be implemented even when as few as two samples per population are available. Using both experimental data and Monte Carlo simulation, we show our approach has improved sensitivity to detect regions enriched for downstream changes in gene expression while accurately controlling the False Discovery Rate (FDR).


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program