duplicate reads still present after MarkDuplicates

[Morgane_AUS]

Hi,

I removed unmapped reads, I ordered the reads and then I tried to remove duplicates using MarkDuplicates but duplicates are still present (same position, same sequence, the only difference is ISIZE, same ISIZE but one is +VE and one is -VE. What does that mean and should they be removed?

Thanks for your help.

Morgane

[MBekritsky]

From the Picard manual page, there is an option called REMOVE_DUPLICATES which can be set to true, false, or null. Did you specify it? Otherwise, the default is NOT to remove duplicates.

Mitch

[Morgane_AUS]

Yes yes I did, it was set on True. True means Remove duplicates right ?

[MBekritsky]

Yeah, that's right...now I'm confused. Can you post a snapshot of the output BAM file?

[dpryan]

I expect that the different isize values you're seeing indicate that they have mates mapped in different places, which would make them not duplicates.

[Morgane_AUS]

How do you paste the data (SAM file) so you are able to scroll to the sides ?

[Brian Bushnell]

Use the "code" tags:

This should scroll if it is long enough.

[Morgane_AUS]

I tried to type <code>mydata </code> but it doesn't work. Also, when I paste some lines from my SAM file, it looses the Tabular format too. Sorry, I'm not sure how to do it. Could you give a bit more details for a newbie please ?

[dpryan]

You can try the "code" or "quote" enclosures. Anyway, I expect we're all familiar enough with SAM files that a little misformatting won't be a big deal.

[Brian Bushnell]

Morgane, you need square brackets: [], not <>