Mapping RNA-seq against genome

Question

[User000]

Hello,

I have RNA-seq data of several varieties from the same plant species. I also have a genome from, let's say, ancestor of that species, which I am going to use as a reference. I am interested in studying SNPs (Single Nucleotide Polimorphisms). Which aligner do you suggest me to use in order to map my RNA-seq data to the reference genome? I am thinking to go for TopHat.

Also, could you suggest some pipeline to do this kind of analyses?

P.S. there is a problem of homoeologous genomes in my plant species

[iraun]

I'd suggest you to give a try to STAR. It is much faster than tophat, though it consumes a considerable amount of RAM. So, having enough computer power, I'd definitely use STAR.

• STAR paper: http://bioinformatics.oxfordjournals.org/content/early/2012/10/25/bioinformatics.bts635
• STAR download page: https://github.com/alexdobin/STAR/releases

[User000]

Thanks for your reply.I have a problem with a computer RAM, basically I want to try to use on a normal computer 1st to see if it works, if no I will have to seek for a server..

[Michele Busby]

Here are the GATK best practices for variant calling in RNA Seq:

http://gatkforums.broadinstitute.org/firecloud/discussion/3891/calling-variants-in-rnaseq

Answer 1

[Devon Ryan]

TopHat is crazy slow. Try STAR (as iraun mentioned), hisat, hisat2, or even BBMap.