RNAseq - removing optical duplicates only

[BENGwins]

Hello all,

I was wondering if anyone knows how to identify and remove only optical duplicates from RNAseq SE data. I have looked into Picard but it was only able to calculate the number of optical duplicates and no option was provided to remove only the optical duplicates. I do not want to remove PCR duplicates especially from single end reads to avoid losing lots of information

Hoping for a good response!

[grosscol]

Use Picard MarkDuplicates

Picard's MarkDuplicates will identify optical duplicates. The option OPTICAL_DUPLICATE_PIXEL_DISTANCE=Integer lets you have control over how finely it will consider the same coordinates to be optical duplicates.

Additionally, the option REMOVE_DUPLICATES=Boolean will allow you to have these marked duplicates included in the output.

Please carefully read the available options and tools at the Picard overview before posting. Often the answer is hiding in the FAQ or Manual.