Generate a VCF File for Different Reference Sequences

Question

[pld]

Is there a means for taking two different reference sequences (e.g. two different virus isolates) and generating a VCF mapping those SNPs? In the sense that I want to see the differences in B with respect to A. I'd like to be able to overlay (in IGV) the differences in a second isolate against the SNPs I found after assembling to the primary isolate of interest.

Answer 1

[Istvan Albert]

In general the VCF format is not suited for describing an alignment between genomes. Nor is IGV a good tool to visualize that.

What you could do however if your really wanted to use both of these is to generate reads from one of your genomes with wgsim, align these reads to the other genome then call SNPs from the alignment.

[pld]

I ended up writing a script that converts a clustal formatted alignment (clustal omega) to VCF. I realize VCF isn't the best tool for this, most of the features of VCF aren't used in this case (depth, quality, etc). The only reason I went with VCF is that it is a format supported by many visualization tools.

The goal was to have something where I can visualize the snps of my isolate of species A with respect to reference A, and visualize the SNPs between reference A and reference B at the same time. IGV is just for me, the end game is to make a figure to show that none of the SNPs in our isolate (species A), match up with SNPs found in species B. I'd like to include a sashimi plot in this to show that the positions where the A-B SNPs are sufficiently covered. There's some concern (unreasonably IMO), that we have some chocolate in our peanut butter.

[lh3]

It would be better to convert your alignment to SAM, which can be visualized, too.

[Coryza]

Hi Joe. Do you have this script available somewhere? :')