Mark duplicate "Picard"

Question

[Nelislam4]

Hi, I am wondering, How mark duplicate of Picard tools works? In other words. How Picard tools detect the duplicated read in BAM file?

coz I actually have replicates of the same RNA-seq samples, and I just want to run differential expression analysis for those samples but I don't know If merging those BAM files and mark the duplicate by Picard tools would solve the issue or not?

Please Help.

Answer 1

[ATpoint]

You should not mark duplicates in RNA-seq, please use the search function on why that is. If you have technical replicates (=same RNA/cDNA sequenced multiple times) you can either merge the BAM files prior to quantification or sum up the columns of the count matrix. DESeq2 afaik has a function to do that as well.

[WouterDeCoster]

In addition to that, it's not because you have technical replicates that you will end up with read duplicates.

[Nelislam4]

Thanks for your help