1000 genomes - a beginner question

[EHC]

Hello,

First, thank you for your time!

I work on yeast NGS. I like to run an analysis based on the 1000 genomes.
What kind of computer/facility I need?

I would like to be able to analyze reads that map to specific 20kb genomic region. I would like to find all reads that at least one of the pair maps this 20kb region and to be able to reassemble them.

Also, if someone can point me to a practical to work with the 1000 genome (what you need to have before you start) it would be great.

Thanks!
EHC

[laura]

The sort of compute facility you need very much depends on what you want to do.

Getting the variants which have been discovered in a particular 20KB region is relatively easy and is doable in a limited disc space (say 500GBytes)

To get the alignments from the bam files for such a region is going to take a lot longer as you have to use samtools to download over ftp your particular 20KB region for more than 1000 individuals and this is both going to time time a disk, I would want to have at least 5TBytes available for your analysis

If you also want all the unmapped reads for the 1094 low coverage individuals you will need 5Tbytes of disc for those files alone before you are getting your subsections for your 20KB region of the genome

Then depending on what sort of assembly tools you wish to use you may need a combination of a lot of compute nodes and at least one machine with a lot of memory

Can you give us details of what you are looking for in these assemblies?

[EHC]

Thank you Laura. I am interested in reads that do not behave as expected (mainly broken reads). Is there any easy way to get these in a specific region.

[laura]

Our FAQ describes how to get subsections of files http://www.1000genomes.org/faq/how-d...ction-bam-file