SNP visualisation

[a11msp]

Hi everyone,

I'm wondering if there's a consensus of what tools to use to visualise SNPs.
I have the data in the form of a SNP table across a number of individuals, and my wet-lab collaborators ask for a map of specific loci, ideally overlaid with genome annotation data.

The SNP table has been assembled from short-read sequencing data
(bowtie assembly -> samtools SNP calling -> samtools pileup per individual -> merger and some additional filtering).

The collaborators really just want to see the SNP positions (and what is there), not the underlying quality/coverage etc.

A quick search for available tools revealed:
- GeneSNP-VISTA (works with FASTA; I can indeed convert my SNP tables into FASTA per individual, but the application didn't start on my Mac)
- Tablet (looks cool, but is designed for quality/coverage formats such as SAM/BAM)

Would appreciate any recommendation, thanks!

[severin]

SNP Visualization using R

I usually find when I am asked for a very specific task that it is easier to just
write an R function to get exactly what you want. Chances are that you already know the SNP coordinates and the sizes of the chromosomes. So it should be relatively straight forward to generate a scaled version of the chromosomes and plot the relative positions. Depending on your needs you could also see if there is any clustering of your SNPs that might indicate a region of introgression (if you have two near isogenic lines). You will want to account for gene density of course. I find a bootstrap algorithm works well for this.

If you want an example of an R script, look at the supplementary materials of
this paper that we just published.

http://www.plantphysiol.org/cgi/content/short/pp.110.158949v1

[adamdeluca]

I think your best bet would be to create custom annotation tracks for the UCSC browser. You could make the mapped reads and the SNPs available to your collaborators and they could turn tracks on and off as they please.

List of comparable formats:

http://genome.ucsc.edu/FAQ/FAQformat.html

I use bed for SNPs, bam for the reads, and bigwig for coverage data.

[a11msp]

Thanks to both of you for the suggestions!
I think I'll follow the route suggested by adamdeluca and convert my SNP tables to a series of bed files