base coverage per position on scaffolds

[boetsie]

Hi all,

i've mapped my reads with Bowtie to a number of scaffolds, and now i want to know on every position on the scaffolds what the coverage is. Does anyone know how to do this, or is there a script/program available?

As output I want something like;

scaffold pos coverage
scaffold1 1 4
scaffold1 2 6
scaffold1 3 10
...
scaffold2 1 5
scaffold2 2 15
...

Thanks in advance,
Marten

[natstreet]

If you produced a sam file using bowtie, you can convert this to bam using samtools and then use bedtools to get per position coverage information.

[drio]

Use samtools pileup:

$ samtools pileup -f your_ref.fa your.bam

Column # 4 is what you wanted.

[boetsie]

Thanks both for your reply. I've used Drio's method since it's the easiest one. It works very good!

[bfantinatti]

Different values

Hello there, I had the same problem, and previous posts solved my problems. But there is an little detail:
I have a bam file, and used mpileup to list the per position coverage. Resulted in a file with some columns where the column #4 is the coverage. But opening in Tablet the same bam file using the same reference genome, the coverage number are not the same. Some one knows what is happening?
Thank you a lot

[whataBamBam]

I'm doing this sort of thing at the moment and just as the following snippet of Perl is really useful for working with Bam files..

#!/usr/bin/perl
open(INFILE,"samtools view $ARGV[0] |");
while(<INFILE>)
{
	#do something with $_;

}

You can also say..

#!/usr/bin/perl
open(INFILE,"samtools mpileup $ARGV[0] |");
while(<INFILE>)
{
	#do something with $_;

}

And the coverage is in column 4. I'm just adding this post to caution that mpileup doesn't report the positions where coverage is zero (my source of this info being some other posts on seqanswers).. but samtools depth does report the zero's.

Does anyone have a nice way to deal with multiply mapping reads here? The standard way to deal with a read that maps to multiple contigs is to count it towards the coverage at every position where it maps right?