Dedup parameters

[gdolsten]

Is there a reason why the default parameters are:

Pairs with both sides mapped within this distance (bp) from each other are considered duplicates. [dedup option] [default: 3]?

In particular, is there a reason why deduplication doesn't remove perfectly identical read pairs?

[Phlya]

I am not sure it has ever been published, one can observe a slight enrichment of pairs within ~3bp of each other, suggesting some steps of the library prep might add or remove 1-3 bp at the ends of fragments. Not sure how widespread it is and how it varies between protocols... That said, it wasn't a huge effect, and the vast majority of duplicates are exact matches, so it's fine to set it to 0.

[gdolsten]

Ok, great thanks. Pairtools keeps one of the representative duplicates though, correct?

[Phlya]

Of course. It keeps the first one it encounters.

[Phlya]

Assume this is solved now, feel free to reopen!