Gatk Pipeline: Markduplicates At The End ?

Question

[Pierre Lindenbaum]

The GATK Mark the Duplicates at the end of their pipeline, after merging the BAMs .

In order to remove the optical duplicates and for each lane, I would have put this operation after the alignment with BWA for each lane/sample (= parallelization = faster)

Is there any reason to mark the duplicates at this position in their pipeline ?

Answer 1

[Stefano Berri]

As far as I understand, here they are sequencing the same LIBRARY in different lanes. I don't know what you mean by "optical duplicates", but what you want to get rid of are PCR duplicate, i.e. the same molecule (produced during PCR amplification) sequenced twice. Either two "spots" on the same lane or in different lanes. That's why you need to mark duplicate after you have merged all reads from a particular library. I guess you could mark duplicate also before, but definetely you need to do it after the merging.

I hope this helps

[Pierre Lindenbaum]

optical duplicate=two spots, close to each other, mapping the same fragment.

[Stefano Berri]

Ok, then those should be very few. PCR duplicates can be many more and more serious. We did have some libraries with up to 70% PCR duplicates. Clearly no good libraries.

[Pierre Lindenbaum]

OK, the "PCR duplicates" is a good argument.