SAMtools rmdup - saving the duplicates?

[KlausNZ]

Is there a way to save the duplicates (write to a separate file) that are removed by SAMtools' rmdup? I'd like to have a look at them....

Many thanks in advance,
K

[volks]

try picard MarkDuplicates and then using samtools view -f 1024

[KlausNZ]

Cheers, that makes sense. I wish there was a way of doing this in a single step with samtools (I'm not so fond of Picard at the moment... )
K

[bruce01]

Is it too obtuse to say that you could grep out the non-matching lines from the rmdup SAM and the original SAM?

Also whats wrong with Picard? I've always found it more consistent.

[robinpsmith]

Picard MarkDuplicates doesn't work with SE reads, as far as I can tell.

We're estimating duplicates in ChIP-seq data.

[volks]

Picard MarkDuplicates doesn't work with SE reads, as far as I can tell.

i am not sure of this. why wouldnt it?

anyway, depending on your alignment file size you could also consider running a unix diff on the original and samtools-rmdup'ed SAM file.

[naragam]

Hi,

Am using samtools rmdup ... but it doesn't seem to remove them in the out file? What's going on?

Nash

[robinpsmith]

@volks you are right, it does work on SE. My bad.

@naragam. I ended up using Picard, it does a much better job.