shinyMethyl: interactive quality control of Illumina 450k DNA methylation arrays in R

Abstract

We present shinyMethyl, a Bioconductor package for interactive quality control of DNA methylation data from Illumina 450k arrays. The package summarizes 450k experiments into small exportable R objects from which an interactive interface is launched. Reactive plots allow fast and intuitive quality control assessment of the samples. In addition, exploration of the phenotypic associations is possible through coloring and principal component analysis. Altogether, the package makes it easy to perform quality assessment of large-scale methylation datasets, such as epigenome-wide association studies or the datasets available through The Cancer Genome Atlas portal. The shinyMethyl package is implemented in R and available via Bioconductor. Its development repository is at https://github.com/jfortin1/shinyMethyl.

Epigenetic signatures of starting and stopping smoking

“…We identified two clusters from DNA methylation data in over 4,900 individuals -one enriched for current smokers and another enriched for never-smokers. It took [15][16][17][18][19] years for the majority of low-dose smokers to display a methylation profile that assigned them to the smoker-enriched cluster. It took less than 1 year for the majority of low-dose exsmokers to be assigned to the never smoker-enriched cluster.…”

“…Quality control was conducted in R [16]. ShinyMethyl [17] was used to plot the log median intensity of methylated versus unmethylated signal per array, with outliers excluded upon visual inspection. WateRmelon [18] was used to remove (1) samples where ≥ 1% of CpGs had a detection p-value in excess of 0.05 (2) probes with a beadcount of less than 3 in more than 5 samples, and (3) probes where ≥ 0.5% of samples had a detection p-value in excess of 0.05.…”

Meffil: efficient normalisation and analysis of very large DNA methylation samples

“…Ideally, then, analysis tools should reflect this, allowing for some level of automation while allowing high-level tasks to be broken down into more specific tasks with customizable solutions. Although it would be most convenient for users to follow and interact with analyses using graphical user interface packages [15,16], such interfaces are often not available on computational servers, particularly when the servers are nodes in a high performance computing cluster. In meffil we address these challenges by providing functions that nearly completely automate the entire process but can be replaced with calls to a sets of functions that allow more detailed interaction with data processing.…”

BoostMe accurately predicts DNA methylation values in whole-genome bisulfite sequencing of multiple human tissues

“…We processed raw signal idat files using minfi v1.20.2 [105] with the Illumina normalization method. We analyzed the quality of each sample looking for outliers across a variety of measures including fraction of failed probes (detection p-value $>$ 0.05), median methylated and unmethylated intensity, control probe signal (using the returnControlStat function from shinyMethyl v1.10.0 [106]), distribution of the overall methylation profile, and principle component analysis. None of the WGBS included in this study were flagged as outliers.…”