Distinguish real duplicates from those generated by PCR

[aquleaf]

Hi All.

Sorry to bother you. Is there a way to distinguish duplicates caused by high sequencing depth from the PCR artefacts?

One more question. Is it necessary to remove the duplicates in the transcription factor ChIP-Seq? In our TF ChIP-Seqs, It is often to see a high duplication level. It is not the case in the controls. I guess the duplicate is caused by high sequencing depth not by the PCR artefacts.

Wish your help! Thanks very much!


Best, Mei

[pbluescript]

Originally Posted by aquleaf

Sorry to bother you. Is there a way to distinguish duplicates caused by high sequencing depth from the PCR artefacts?

Only if you added a unique barcode to each individual molecule during library prep before PCR. There are a couple papers out there that have done this. Here's one example:
http://nar.oxfordjournals.org/conten...ar.gkr217.long

One more question. Is it necessary to remove the duplicates in the transcription factor ChIP-Seq? In our TF ChIP-Seqs, It is often to see a high duplication level. It is not the case in the controls. I guess the duplicate is caused by high sequencing depth not by the PCR artefacts.

I would not remove any potential duplicates from ChIP-Seq data. You are sampling from a smaller portion of the genome, so duplicates are expected.