plotCorr: Show DE/DM state correlation between cell types

In ziyili20/TOAST: Tools for the analysis of heterogeneous tissues

Show DE/DM state correlation between cell types

Description

This function generates -log10 transformed p-values for each pair of cell types and calculate corresponding Pearson correlation and odds ratio - a modification of function 'ggpairs' in package 'GGally'.

Usage

plotCorr(pval, de.state=NULL, pval.thres=NULL, fdr.thres=NULL, p.size = 0.2,
         p.color = grDevices::adjustcolor( "black", alpha.f = 0.2),
         fig.margin = c(1,1,1,1),
         fig.margin.unit = 'in', line.type = 'dashed', line.color = 'blue')

Arguments

pval	matrix of p-values, with rows representing features and columns representing cell types.
de.state	(optional) matrix of DE/DM states (1: DE/DM, 0:non-DE/DM), with rows representing features and columns representing cell types.
pval.thres	threshold of p-value to define DE/DM, required if de.state not provided.
fdr.thres	threshold of FDR to define DE/DM, required if de.state not provided and FDR is prefered than p-value.
p.size	point size for scatter plot
p.color	point color for scatter plot
fig.margin	figure margin
fig.margin.unit	unit of figure margin
line.type	line type in scatter plot
line.color	line color in scater plot

Value

A figure contains scatter plots, Pearson correlaiton and odds ration of -log10 transformed p-values for each pair of cell types.

Author(s)

Luxiao Chen <[e-mail]>

Examples

pval.1 <- runif(1000,0,1)
pval.2 <- pval.1 + rnorm(1000,0,0.01)
pval.2[pval.2 < 0] =0
pval.3 <- runif(1000,0,1)

pval.input <- data.frame('cell.1'=pval.1,
                         'cell.2'=pval.2,
                         'cell.3'=pval.3)

plotCorr(pval = pval.input, pval.thres = 0.05)

ziyili20/TOAST documentation built on Aug. 28, 2022, 11:28 a.m.