mqtl.md

In mw201608/NetWeaver: Graphic Presentation of Complex Genomic and Network Data Analysis

NetWeaver

Trans-mQTLs in the developing human brain

Here I will demonstrate how to use NetWeaver to make a typical circos plot illustrating the genome-wide DNA methylation quantitative trait loci (mQTLs). The mQTLs dataset that I am going to use was from Hannon et al (PMID 2661935) which studied the human fetal brain DNA methylation levels. I downloaded the imputed set of mQTLs at link https://epigenetics.essex.ac.uk/mQTL/. There are 658492 SNP-methylation probe pairs in this dataset. For simplicity, only the top 1000 most significant pairs will be used in this demonstration. You may load this preprocessed mQTLs data by:

library(NetWeaver)
load(mqtl)
head(mqtl)

A typical circos plot in the genomics study will generally show chromosome cytoband information in the out-most track. For convenience, two human chromosome cytobands for genome build hg19 and hg38 obtained from the UCSC genome browser have been included in the NetWeaver pakcage. In this tutorial, we will use hg19 cytoband which can be imported by

data(ucsc.hg19.cytoband)

Note the hg38 version cytoband can be imported by data(ucsc.hg38.cytoband).

Now, let us initialize the circos plot parameters and prepare the canvas:

rc.initialize(ucsc.hg19.cytoband, num.tracks=16, params=list(chr.padding=0.1))
rc.plot.area(size=0.9)

Plot ideogram and chromosome ids at track 1 and 2

rc.plot.ideogram(track.id=1:2, plot.band=TRUE, plot.chromosome.id=TRUE)

Prepare link data from mQTLs:

LinkData=data.frame(Chr1=paste0('chr',mqtl$SNP_Chr), Pos1=mqtl$SNP_BP, 
    Chr2=paste0('chr',mqtl$DNAm_CHR),Pos2=mqtl$DNAm_BP, Data=abs(mqtl$beta),stringsAsFactors=FALSE)
````

We will color the links by the chromosomal location of the mQTL SNPs.

chromosomes=paste0('chr',c(1:22,'X','Y')) chr.factor2color=rlib:::labels2colors(chromosomes) LinkData$Color=chr.factor2color[as.integer(factor(LinkData$Chr1,levels=chromosomes))]

Plot color bar for each chromosome.

chr.color=do.call(rbind,lapply(split(ucsc.hg19.cytoband,ucsc.hg19.cytoband$Chr),function(x) data.frame(Chr=x$Chr[1],Start=1,End=max(x$End),stringsAsFactors=FALSE))) chr.color$Color=chr.factor2color[as.integer(factor(chr.color$Chr,levels=chromosomes))] rc.plot.barchart(chr.color, track.id=1, color.col="Color", custom.track.height=rc.get.params()$track.height/3, track.border=NA)

Lastly, plot the links:

rc.plot.link(LinkData, track.id=4, data.col=5,color.col=6)

mw201608/NetWeaver documentation built on Aug. 28, 2023, 6:45 p.m.