merging a tab and a fasta file

[arg]

Hi all,
I recently started to use Illumina and now that I have tons of data I need to deal with them. I downloaded the Nematostella genome to blast my contigs against it, but it seems that the actual sequences and the annotations are in separate files (actually fasta and tab files respectively). I have no idea how to merge them, but I really need to figure this out. Can anybody help me? THanks!!!!
Ana

[mrawlins]

Using blast is likely a terrible idea if you have tons of data. It just isn't fast enough. A wide array of software have been designed to do match your Illumina reads to the genome. My personal favorite (after attempting to use BFAST, BWA and BioScope) is Bowtie. The Bowtie/TopHat/Cufflinks pipeline is very popular for this sort of thing.

What you want to do first is map your reads to the genome (which comes in the fasta file). This fasta should include all the contigs for all chromosomes and/or plasmids. That way it will map all the intergenic reads properly. The tab file with annotations comes later for quantifying how many reads mapped to each gene. I wrote a little java program to do this, but I'm pretty sure TopHat/Cufflinks will do this for you. Merging the sequences and annotations before mapping will result in losing all information from intergenic sequences which can provide quite a bit of biologically relevant data, so I wouldn't recommend it.

If you're not comfortable programming I suggest TopHat/Cufflinks. There are quite a few discussions on using them already in this forum, as well as some pretty good tutorials on the website.

Good luck, and welcome to the community!

[arg]

Hey mrawlins,
thanks for your quick response. I will try Bowtie right away...I hope I can understand it

Using blast is likely a terrible idea if you have tons of data. It just isn't fast enough. A wide array of software have been designed to do match your Illumina reads to the genome. My personal favorite (after attempting to use BFAST, BWA and BioScope) is Bowtie. The Bowtie/TopHat/Cufflinks pipeline is very popular for this sort of thing.

What you want to do first is map your reads to the genome (which comes in the fasta file). This fasta should include all the contigs for all chromosomes and/or plasmids. That way it will map all the intergenic reads properly. The tab file with annotations comes later for quantifying how many reads mapped to each gene. I wrote a little java program to do this, but I'm pretty sure TopHat/Cufflinks will do this for you. Merging the sequences and annotations before mapping will result in losing all information from intergenic sequences which can provide quite a bit of biologically relevant data, so I wouldn't recommend it.

If you're not comfortable programming I suggest TopHat/Cufflinks. There are quite a few discussions on using them already in this forum, as well as some pretty good tutorials on the website.

Good luck, and welcome to the community!