iplotCorr {qtlcharts}

Image of correlation matrix with linked scatterplot

Description

Creates an interactive graph with an image of a correlation matrix linked to underlying scatterplots.

Usage

iplotCorr(
  mat,
  group = NULL,
  rows = NULL,
  cols = NULL,
  reorder = FALSE,
  corr = NULL,
  scatterplots = TRUE,
  chartOpts = NULL,
  digits = 5
)

Arguments

mat

Data matrix (individuals x variables)

group

Optional vector of groups of individuals (e.g., a genotype)

rows

Selected rows of the correlation matrix to include in the image. Ignored if 'corr' is provided.

cols

Selected columns of the correlation matrix to include in the image. Ignored if 'corr' is provided.

reorder

If TRUE, reorder the variables by clustering. Ignored if 'corr' is provided as a subset of the overall correlation matrix

corr

Correlation matrix (optional).

scatterplots

If ‘FALSE', don’t have the heat map be linked to scatterplots.

chartOpts

A list of options for configuring the chart (see the coffeescript code). Each element must be named using the corresponding option.

digits

Round data to this number of significant digits before passing to the chart function. (Use NULL to not round.)

Details

'corr' may be provided as a subset of the overall correlation matrix for the columns of 'mat'. In this case, the 'reorder', 'rows' and 'cols' arguments are ignored. The row and column names of 'corr' must match the names of some subset of columns of 'mat'.

Individual IDs are taken from 'rownames(mat)'; they must match 'names(group)'.

Value

An object of class 'htmlwidget' that will intelligently print itself into HTML in a variety of contexts including the R console, within R Markdown documents, and within Shiny output bindings.

See Also

[iheatmap()], [scat2scat()], [iplotCurves()]

Examples

data(geneExpr)

iplotCorr(geneExpr$expr, geneExpr$genotype, reorder=TRUE,
          chartOpts=list(cortitle="Correlation matrix",
                         scattitle="Scatterplot"))

# use Spearman's correlation
corr <- cor(geneExpr$expr, method="spearman", use="pairwise.complete.obs")
# order by hierarchical clustering
o <- hclust(as.dist(1-corr))$order

iplotCorr(geneExpr$expr[,o], geneExpr$genotype, corr=corr[o,o],
          chartOpts=list(cortitle="Spearman correlation",
                         scattitle="Scatterplot"))