sEst

An R-package for estimating the sex from DNA methylation microarray data

Installation

1) Click on the package source file (‘sest.tar’).

2) Click on the ‘Download’ button.

3) Install the package from source:

a) In command line:

R CMD INSTALL '/path/to/package source'

b) Or, in R environment:

install.packages(pkgs="/path/to/package source", repos=NULL, type="source")

Alternatively:

install.packages(pkgs="https://github.com/jungch/sest/raw/master/sest.tar", repos=NULL, type="source")

Example codes

Here is an example scenario of estimating the sex of HM450k data samples. This example uses a publicly available dataset from GEO (GSE51032).

Functions:

‘estimateSex’

usage:

estimateSex(beta.value = NULL, detecP = NULL,
  beta.intervals.X = seq(0, 1, 0.1),
  beta.intervals.Y = seq(0, 1, 0.1),
  p.intervals.X = c(-18, -5, -2, 0),
  p.intervals.Y = c(-18, -5, -2, 0),
  return_with_reference = FALSE)

Arguments:

name | description

• beta-value	beta-value matrix
  detecP	detection p-value matrix
  beta.intervals.X	Intervals of beta-value of X chromosome probes within each of which the proportion is calculated. Default is ‘seq(0,1,0.1)’, which is 0 to 0.1, 0.1 to 0.2, …, 0.9 to 1.0.
  beta.intervals.Y	Intervals of beta-value of Y chromosome probes within each of which the proportion is calculated. Default is the same as that of ‘beta.intervals.X’
  p.intervals.X	Intervals of detection p-value of X chromosome probes within each of which the proportion is calculated. Default is -Inf to 1e-18, 1e-18 to 1e-5, 1e-5 to 1e-2, 1e-2 to 0.
  p.intervals.Y	Intervals of detection p-value of Y chromosome probes within each of which the proportion is calculated. Default is the same as that of ‘p.intervals.X’
  return_with_reference	Logical value to indicate whether or not to return the data derived from references data (Default to FALSE).

‘plotSexEstimation’

Description:

This function generates a scatter plot using ‘ggplot’ showing the first principal component of two principal component analysis results (pincipal component analysis using the data from X chromosome and the data from Y chromosome)

Usage:

plotSexEstimation(sex_estimation = NULL, samples = NULL,
  main = "", filename = NULL,
  include_reference = FALSE)

Arguments:

name | Description

• sex_estimation	result of estimateSex function
  samples	the list of samples to be plotted. If NULL (default), all samples in the estimateSex result are plotted.
  main	character string to be used as the plot title. Default is an empty string.
  filename	When specified, a PDF is generated.
  include_reference	Logical value to specify whether to plot the reference data (Default is FALSE).

Examples:

plotSexEstimation(sex_estimation = sexEst)

‘plotSexDistribution’

Description:

This function generates a line chart using ‘ggplot’ showing the distribution of beta-values of chrX and chrY and detection p-values of chrY.

Usage:

plotSexDistribution(beta.value = NULL, detecP = NULL,
  beta.intervals.X = seq(0, 1, 0.1),
  beta.intervals.Y = seq(0, 1, 0.1),
  p.intervals.X = c(-18, -5, -2, 0),
  p.intervals.Y = c(-18, -5, -2, 0),
  samples = NULL, filename = NULL,
  include_reference = TRUE,
  color = NULL)

Arguments:

name | description

• beta-value	beta-value matrix
  detecP	detection p-value matrix
  beta.intervals.X	Intervals of beta-value of X chromosome probes within each of which the proportion is calculated. Default is ‘seq(0,1,0.1)’, which is 0 to 0.1, 0.1 to 0.2, …, 0.9 to 1.0.
  beta.intervals.Y	Intervals of beta-value of Y chromosome probes within each of which the proportion is calculated. Default is the same as that of ‘beta.intervals.X’
  p.intervals.X	Intervals of detection p-value of X chromosome probes within each of which the proportion is calculated. Default is -Inf to 1e-18m 1e-18 to 1e-5, 1e-5 to 1e-2, 1e-2 to 0.
  p.intervals.Y	Intervals of detection p-value of Y chromosome probes within each of which the proportion is calculated. Default is the same as that of ‘p.intervals.X’
  samples	the list of samples to be plotted. If NULL (default) all samples in the beta.value matrix are plotted.
  filename	When specified, a PDF is generated.
  include_reference	Logical value to specify whether to plot reference data (Default is FALSE).
  color	Logical value to specify whether to plot the reference data (Default is ‘#984ea3’, one of the shades of purple).
  main	character string to be used as the plot title. Default is an empty string.

Examples:

plotSexDistribution(beta.value=beta, detecP=pval)

‘get.proportion_table’

Description:

This function returns the beta-value and detection p-value proportion table within each intervals specified by parameters.

Usage:

get.proportion_table(beta.value = NULL, detecP = NULL,
   beta.intervals.X = seq(0, 1, 0.1),
   beta.intervals.Y = seq(0, 1, 0.1),
   p.intervals.X = c(-18, -5, -2, 0),
   p.intervals.Y = c(-18, -5, -2, 0),
   samples = NULL)

Arguments:

name | description

• beta-value	beta-value matrix
  detecP	detection p-value matrix
  beta.intervals.X	Intervals of beta-value of X chromosome probes within each of which the proportion is calculated. Default is ‘seq(0,1,0.1)’, which is 0 to 0.1, 0.1 to 0.2, …, 0.9 to 1.0.
  beta.intervals.Y	Intervals of beta-value of Y chromosome probes within each of which the proportion is calculated. Default is the same as that of ‘beta.intervals.X’
  p.intervals.X	Intervals of detection p-value of X chromosome probes within each of which the proportion is calculated. Default is -Inf to 1e-18m 1e-18 to 1e-5, 1e-5 to 1e-2, 1e-2 to 0.
  p.intervals.Y	Intervals of detection p-value of Y chromosome probes within each of which the proportion is calculated. Default is the same as that of ‘p.intervals.X’
  samples	list of samples to calculate beta/detection p-value proportions. If NULL (default) all samples in the beta-value and detection p-value matrices are used.

Examples:

proportion_table <- get.proportion_table(beta.value=beta,
  detecP=detecP)