Search

This chapter describes methods currently available for visualizing results from systems genetics experiments. Here, we abstract from the statistical methods used for genetic mapping, which are dependent on the specific resource being used, i.e. F2, RILs, or outbred populations among others. We use a public dataset with results from a mouse eQTL experiment for three examples of visualization: genome-wide dot plots of marker-by-gene association, karyotype-like plots, and circos plots. Dot plots give a first overview of the results from eQTL mapping, allowing detecting genome-wide patterns of cis- and trans-genetic association to transcription level. Karyotype-like plots provide chromosomal context and allow integrating multiple tracks of information in a single plot. Circos plots can, in addition, display long-range interactions to provide an overview of genetic connectivity at the genome level. All examples are developed and explained using R code, an open-source language with powerful statistical and graphical capabilities. The principles reviewed here, however, can be applied with other software options, organisms, and to any type of molecular phenotype that can be assigned to a genomic position.

This chapter describes methods currently available for visualizing results from systems genetics experiments. Here, we abstract from the statistical methods used for genetic mapping, which are dependent on the specific resource being used, i.e. F2, RILs, or outbred populations among others. We use a public dataset with results from a mouse eQTL experiment for three examples of visualization: genome-wide dot plots of marker-by-gene association, karyotype-like plots, and circos plots. Dot plots give a first overview of the results from eQTL mapping, allowing detecting genome-wide patterns of cis- and trans-genetic association to transcription level. Karyotype-like plots provide chromosomal context and allow integrating multiple tracks of information in a single plot. Circos plots can, in addition, display long-range interactions to provide an overview of genetic connectivity at the genome level. All examples are developed and explained using R code, an open-source language with powerful statistical and graphical capabilities. The principles reviewed here, however, can be applied with other software options, organisms, and to any type of molecular phenotype that can be assigned to a genomic position.

Summary: Visualizing genomic data in chromosomal context can help detecting errors in data generation or analysis and can suggest new hypotheses to be tested. Here we report a new tool for displaying large and diverse genomic data in idiograms of one or multiple chromosomes. The package is implemented in R so that visualization can be easily integrated with its numerous packages for processing genomic data. It supports simultaneous visualization of multiples tracks of data, each of potentially different nature. Large genomic regions such as QTLs or synteny tracts may be shown along histograms of number of genes, genetic variants, or any other type of genomic element. Tracks can also contain values for continuous or categorical variables and the user can choose among points, points connected by lines, line segments, barplots or histograms for representing data. chromPlot reads data from tables in BED format which are imported in R using its builtin functions. The information necessary to draw chromosomes for mouse and human is included with the package. Chromosomes for other organisms are downloaded automatically from the Ensembl website or can be provided by the user. We present common use cases here, and a full tutorial is included as the packages's vignette.Availability: chromPlot is distributed under a GLP2 licence at Genomed Lab: http://genomed.med.uchile.cl.Contact:raverdugo@u.uchile.cl