kpPlotManhattan: kpPlotManhattan

Description Usage Arguments Details Value See Also Examples

View source: R/kpPlotManhattan.R

Description

Creates a manhattan plot, the ones usually seen in GWAS studies.

Usage

1
2
3
4
5
6
kpPlotManhattan(karyoplot, data=NULL, pval=NULL, points.col="2grays", 
points.pch=16, points.cex=1, suggestiveline = -log10(1e-05), 
suggestive.col="black", suggestive.lwd=1, suggestive.lty=2, 
genomewideline = -log10(5e-08), genomewide.col="black", genomewide.lwd=1, 
genomewide.lty=1, logp=TRUE, highlight=NULL, highlight.col="greenyellow",
ymin=NULL, ymax=NULL, data.panel=1, r0=NULL, r1=NULL, clipping=TRUE, ...)

Arguments

karyoplot

(a KaryoPlot object) This is the first argument to all data plotting functions of karyoploteR. A KaryoPlot object referring to the currently active plot.

data

(a GRanges) A GRanges object (or any other format accepted by toGRanges) with the data points to plot. If the pval parameter is NULL and data has a column named "p" or "pval", it will be used as the pvalues to plot.

pval

(numeric) The pvalues to plot. It must have the same length as data. If NULL, data$p or data$pval will be used, if present. (defaults to NULL)

points.col

(colors or character) The colors used to plot the points. It can be either a vector of colors of the same length of data or a valid color specification for colByChr. (defaults to "2grays")

points.pch

(numeric between 1 and 25) The symbol used to plot every point. (Defaults to 16, a filled circle)

points.cex

(numeric) The size of the point symbols. (defaults to 1)

suggestiveline

(numeric) The suggestive significance threshold. The suggestive line will be plotted at this vertical position. If NULL, no line will be plotted. (defaults to -log10(1e-05))

suggestive.col

(color) The color of the suggestive line. (defaults to "black")

suggestive.lwd

(numeric) The width of the suggestive line (defaults to 1)

suggestive.lty

(numeric) The line type of the suggestive line (defaults to 2, dashed line)

genomewideline

(numeric) The genomewide significance threshold. The genomewide line will be plotted at this vertical position. If NULL, no line will be plotted (defaults to -log10(5e-08))

genomewide.col

(color) The color of the genomewide line. (defaults to "black")

genomewide.lwd

(numeric) The width of the genomewide line. (defaults to 1)

genomewide.lty

(numeric) The line type of the genomewide line. (defaults to 1, solid line)

logp

(logical) If TRUE, pval will be transformed using -log10(pval). (defaults to TRUE)

highlight

(GRanges, character vector, logical vector or numeric vector) The points to highlight in a different color. If a GRanges (or anythng accepted by toGRanges) the points overlapping these regions will be highlighted. Otherwise the points will be selected with data[highlight]. If NULL no point will be highlighted. (defaults to NULL)

highlight.col

The color of the highlighted points (defaults to "greenyellow")

ymin

(numeric) The minimum value to be plotted on the data panel. If NULL, it is set to 0. (deafults to NULL)

ymax

(numeric) The maximum value to be plotted on the data.panel. If NULL, it is set to 1. (defaults to NULL)

data.panel

(numeric) The identifier of the data panel where the data is to be plotted. The available data panels depend on the plot type selected in the call to plotKaryotype. (defaults to 1)

r0

(numeric) r0 and r1 define the vertical range of the data panel to be used to draw this plot. They can be used to split the data panel in different vertical ranges (similar to tracks in a genome browser) to plot differents data. If NULL, they are set to the min and max of the data panel, it is, to use all the available space. (defaults to NULL)

r1

(numeric) r0 and r1 define the vertical range of the data panel to be used to draw this plot. They can be used to split the data panel in different vertical ranges (similar to tracks in a genome browser) to plot differents data. If NULL, they are set to the min and max of the data panel, it is, to use all the available space. (defaults to NULL)

clipping

(boolean) Only used if zooming is active. If TRUE, the data representation will be not drawn out of the drawing area (i.e. in margins, etc) even if the data overflows the drawing area. If FALSE, the data representation may overflow into the margins of the plot. (defaults to TRUE)

...

The ellipsis operator can be used to specify any additional graphical parameters. Any additional parameter will be passed to the internal calls to the R base plotting functions.

Details

Creates a manhattan plot, the ones usually seen in GWAS studies. By default, it will compute the -log10 of the pvalues given and plot them as points. In addition, it can plot to horizontal lines, one for the "suggestive" threshold and another one for the "genomewide" significance threshold. In addition, it can highlight some of the data points in a different color. Highlighted data points can be specified per name, per position in the data structure or by their position on the genome (see examples).

There's more information at the https://bernatgel.github.io/karyoploter_tutorial/karyoploteR tutorial.

Value

Returns the original karyoplot object with the data computed (ymin, ymax, suggestiveline, genomewideline and data (with two additional columns pval and color)) stored at karyoplot$latest.plot

See Also

plotKaryotype, kpPoints, colByChr, colByRegion

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
set.seed(1000)

#First simulate a GWAS result with a single significant peak
data <- createRandomRegions(nregions=20000, length.mean=1, length.sd=0, genome=filterChromosomes(getGenome("hg19")))
names(data) <- paste0("rs", 1:20000)
data$pval <- rnorm(n = 20000, mean = 0.5, sd = 0.5)
data$pval[data$pval<0] <- -1*data$pval[data$pval<0]
snps.in.peak <- which(overlapsAny(data, toGRanges("chr3:70e6-80e6")))
data$pval[snps.in.peak] <- runif(n = length(snps.in.peak), min=0.1, max=8)
data$pval <- 10^(-1*data$pval)
 
kp <- plotKaryotype("hg19", plot.type=4)
kp <- kpPlotManhattan(kp, data, ymax=8)
kpAxis(kp, ymax=8)

#Highlighting 
kp <- plotKaryotype("hg19", plot.type=4)
kp <- kpPlotManhattan(kp, data, ymax=8, highlight="chr3:70e6-80e6", r0=autotrack(1,4))
kp <- kpPlotManhattan(kp, data, ymax=8, highlight=snps.in.peak, highlight.col="orchid", r0=autotrack(2,4))
kp <- kpPlotManhattan(kp, data, ymax=8, highlight=names(data)[snps.in.peak], highlight.col="orange", r0=autotrack(3,4))
kp <- kpPlotManhattan(kp, data, ymax=8, highlight=overlapsAny(data, toGRanges("chr3:70e6-80e6")), r0=autotrack(4,4))

#Look and feel
kp <- plotKaryotype("hg19", plot.type=4)
kp <- kpPlotManhattan(kp, data, ymax=8, points.col="2blues", highlight="chr3:70e6-80e6", points.pch=2, points.cex=0.6, suggestive.col="red", suggestive.lwd=3, suggestive.lty=4)

karyoploteR documentation built on Nov. 8, 2020, 5:52 p.m.