A Bayesian Approach for Analysis of Whole-Genome Bisulfite Sequencing Data Identifies Disease-Associated Changes in DNA Methylation

Abstract

Whole-genome bisulphite sequencing (WGBS) can identify important methylation differences between diseased and healthy samples. However, results from...

Identification of Ceruloplasmin as a Gene that Affects Susceptibility to Glomerulonephritis Through Macrophage Function

“…During the development of crescentic glomerulonephritis, the major pathogenic event that causes crescent formation is the rupture of glomerular capillaries, which allows a relatively early macrophage infiltration into the Bowman’s space. There have been numerous reports showing that macrophage activity and numbers are critical in the inflammatory phase of Crgn ( Duffield et al 2005 ; Wang et al 2007 ; Wang and Harris 2011 ) and our group has contributed to the identification of genetic and epigenetic determinants of macrophage function, which associate with susceptibility to Crgn in rats and humans ( Aitman et al 2006 ; Behmoaras et al 2008 , 2010 ; Page et al 2012 ; Deplano et al 2013 ; Hull et al 2013 ; Kang et al 2014 ; Rackham et al 2017 ).…”

MethylCal: Bayesian calibration of methylation levels

“…The crossed random-effects AMP i and CpG j follow a normal distribution, AMP\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$_{i}|\tau \sim \operatorname{N}(0,\tau ^{-1})$\end{document} and CpG\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$_{j}|\upsilon \sim \operatorname{N}(0,\upsilon ^{-1})$\end{document} with a non-informative prior precision \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\tau \sim \operatorname{Gam}(1,0.1)$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$(\operatorname{E}(\tau ) = 10$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\operatorname{Var}(\tau ) = 100)$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\upsilon \sim \operatorname{Gam}(1,0.1)$\end{document}. For the LGF, we follow (24) and model \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\boldsymbol{\mu }$\end{document} as a Random Walk of order 1 (RW1) (25), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\mu _{j} | \mu _{j-1}, \rho _{j} \sim \operatorname{N}(\mu _{j-1},\rho _{j})$\end{document}, where ρ j = ρ| p j − p j − 1 | with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\rho ^{-1} \sim \operatorname{Gam}(1,0.1)$\end{document} and p j and p j − 1 the chromosomal position of two consecutive CpGs (with p 0 = 0). With this specification the dependence between methylation levels depends on the distance between the corresponding CpGs, i.e., the closer the CpGs, the stronger the dependence.…”

Review of applications of high-throughput sequencing in personalized medicine: barriers and facilitators of future progress in research and clinical application

“…Bisulphite conversion is necessary for both older (microarray) and newer sequencing-based technologies, facilitating the detection of methylated cytosines (one of four DNA component bases), though is a harsh process that can affect the quality of DNA for downstream analysis [37]. Whole-genome bisulphite sequencing (WGBS) was used to identify methylation of the IFITM3 gene as a candidate in the development of kidney disease [38]. Legendre et al [39] used WGBS to develop a blood-based methylation patterns that could be used to stratify breast cancer patients into metastatic disease risk groups.…”