Hereditary Genetics: Current Research

Robust joint statistical tests for DNA methylation data

4^th International Congress on Epigenetics & Chromatin

September 03-05, 2018 | London, UK

Xiaogang Steven Wang

York University, Canada

Scientific Tracks Abstracts: Hereditary Genet Curr Res

Abstract:

The detection of differential methylation between the control and case group of a disease of interest remains a challenging task. Reliable result depends on a correct employment of robust statistical tests. In the past, the hypothesis testing for methylation data is mainly focused on the equality of means while the intrinsic variability in the methylation data has often been ignored. Many statistical tests that could incorporate the variability into the testing procedure rely on the assumption of either normality or absence of outliers. Consequently, these tests are not very robust when applied to real data. In this talk, we will present three robust statistical tests that can test the equality of both means and variances simultaneously. Results from simulation studies and real data analyses suggest that our proposed methods are more powerful and robust in comparison with other three bench mark testing procedures. Recent Publications 1. Gold N, Frasch M, Herry C, Richardson B and Wang X (2018) A doubly stochastic change point detection algorithm for noisy biological signals. Frontiers in Computational Physiology and Medicine 8:1112. 2. Li X, Qiu W, Morrow J, Weiss J, Fu Y and Wang X (2015) A comparative study of tests for homogeneity of variances with application of DNA methylation data. PLOS One 10(12):e0145295. 3. Wang Q, Gold N, Frasch M, Huang H ?? and Wang X (2015) Mathematical model of cardiovascular and metabolic responses to umbilical cord occlusions in fetal sheep. Bulletin of Mathematical Biology 77(12):2264-2293.

Biography :