PlotCorrHeatMap: Plot correlations heatmap

In flajole/MApckg: R package of MetaboAnalyst functions

Description

Plot correlations heatmap and write an output file "correlation_table.csv". Note, the heatmap will only show correlations for a maximum of 1000 features. For larger datasets, only top 1000 features will be selected based on their interquantile range (IQR). When color distribution is fixed, you can potentially compare the correlation patterns among different data sets. In this case, you can choose "do not perform clustering" for all data set, or only to perform clustering on a single reference data set, then manually re-arranged other data sets according to the clustering pattern of the reference data set.

Usage

PlotCorrHeatMap(dataSet, analSet, imgName = "corr_heat_", format = "png",
  dpi = 72, width = NA, cor.method = "pearson", colors = "default",
  viewOpt = "overview", fix.col = F, no.clst = FALSE, top = F,
  top.num = 999)

Arguments

dataSet	List, data set object generated by MS_to_MA function.
analSet	List, containing the results of statistical analysis (can be just an empty list).
imgName	Image file name prefix.
format	Image format, one of: "png", "tiff", "pdf", "ps", "svg"
dpi	Image resolution.
width	Image width.
cor.method	Method of correlation calculation, one of "pearson", "kendall", "spearman"
colors	The color contrast. One of "default", "gbr" (red/green), "heat", "topo", "gray"
viewOpt	View mode, "overview" or "detailed"
fix.col	If TRUE, color distribution is fixed.
no.clst	If TRUE, no clustering lines are plotted.
top	If TRUE, use total abs(correlation) in selection.
top.num	The number of the features with the best contrast to be plotted.

See Also

PatternHunter, PlotCorr

flajole/MApckg documentation built on May 16, 2019, 1:16 p.m.