glimmaMA.MArrayLM: Glimma MA Plot

Description Usage Arguments Details Value See Also Examples

View source: R/glimmaMA.R

Description

Draws a two-panel interactive MA plot from an MArrayLM object. This is a special case of the glimmaXY plot.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
## S3 method for class 'MArrayLM'
glimmaMA(
  x,
  dge = NULL,
  counts = dge$counts,
  groups = dge$samples$group,
  coef = ncol(x$coefficients),
  status = limma::decideTests(x),
  anno = x$genes,
  display.columns = NULL,
  status.cols = c("dodgerblue", "silver", "firebrick"),
  sample.cols = NULL,
  p.adj.method = "BH",
  transform.counts = c("logcpm", "cpm", "rpkm", "none"),
  main = colnames(x)[coef],
  xlab = "logCPM",
  ylab = "logFC",
  html = NULL,
  width = 920,
  height = 920,
  ...
)

Arguments

x

MArrayLM object from which summary statistics are extracted from to create summary (left) plot.

dge

DGEList object with nrow(x) rows from which expression values are extracted from to create expression (right) plot. Gene counts are taken from dge$counts and sample groups from dge$samples$group.

counts

numeric matrix with nrow(x) rows containing gene expression values. This can be used to replace raw gene counts from dge$counts with transformed counts e.g. logCPM or logRPKM values.

groups

vector of length ncol(dge) representing categorisation of samples in expression plot.

coef

integer indicating the column in x from the summary plot is created.

status

vector of length nrow(x) indicating the status of each gene. By default genes in the summary plot are coloured based on its differential expression status using an adjusted p-value cutoff of 5% by calling the limma::decideTests function, where the value of -1 marks down-regulated genes, 0 marks genes with no expression difference, and 1 marks up-regulated genes.

anno

dataframe with nrow(x) rows containing gene annotations.

display.columns

character vector containing names of columns from anno from which to display in mouseover tooltips and table.

status.cols

vector of length 3 containing valid CSS strings for colours associated with status in the order of -1, 0 and 1.

sample.cols

character vector of length ncol(counts) containing valid CSS strings for colours associated with each sample to be displayed on the expression plot. If left unspecified, samples will be coloured according to groups.

p.adj.method

character string specifying p-value adjustment method.

transform.counts

the type of transform used on the counts log-cpm by default. edgeR::cpm(counts, log=TRUE); defaults to FALSE.

main

character string for the main title of summary plot.

xlab

character string for the x-axis label of summary plot.

ylab

character string for the y-axis label of summary plot.

html

character string for naming HTML file for exportation of widget. The extension should be included in the file name e.g. "file.html".

width

numeric value indicating width of widget in pixels.

height

numeric value indicating width of height in pixels.

...

addition unused arguments.

Details

The summary plot on the left represents gene-wise log-fold-change (logFC) on the y-axis versus average gene expression calculated as log-counts-per-million (logCPM) values. We call our summary plot an MA plot because this type of plot was originally referred to as an MA plot in the limma package, with the M-value representing logFC and A-value representing average expression - it has since been renamed to MD plot in the limma package. The expression plot on the right displays sample expression values for a single gene. Interactions with the htmlwidget include clicking on genes (points) in the summary plot to bring up associated sample expression values in the expression plot, as well as the summary statistics in the table below. Alternatively, users can interact with the table by clicking on genes (rows) to highlight genes in the summary plot, as well as bring up associated sample expression values in the expression plot. Briefly, other interactive features include a search box for the table, buttons to save plots and data (summary statistics and expression values), additional pop-up information when hovering on points in plots, and rescaling of the y-axis in the expression plot.

Value

htmlwidget object or NULL if html argument is specified.

See Also

glimmaMA, glimmaMA.DGEExact, glimmaMA.DGELRT, glimmaMA.DESeqDataSet

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
dge <- readRDS(
  system.file("RNAseq123/dge.rds", package = "Glimma"))
design <- readRDS(
  system.file("RNAseq123/design.rds", package = "Glimma"))
contr.matrix <- readRDS(
  system.file("RNAseq123/contr.matrix.rds", package = "Glimma"))

v <- limma::voom(dge, design)
vfit <- limma::lmFit(v, design)
vfit <- limma::contrasts.fit(vfit, contrasts = contr.matrix)
efit <- limma::eBayes(vfit)

glimmaMA(efit, dge = dge)

Glimma documentation built on Nov. 8, 2020, 6:13 p.m.