Any pitfalls to consider when using MACS on paired-end data when forward and reverse reads are different lengths?

Question

[rleach]

I have some paired-end data where the forward and reverse reads are different lengths (due to unrelated requirements of someone else's samples on a multiplexed sequencing run).

Can I use the entirety of both reads when I predict peaks with MACS2 or should I trim the longer reverse reads? Are there any parameters that should change to account for the different read lengths?

Are there any potential problems from using the different length forward/reverse reads?

Answer 1

[ATpoint]

There should not be any problems. The fragment length when using -f BAMPE is defined as the number of nucleotides between the 5' ends of the two reads, so it does not matter how long they are.

[rleach]

Great. Thanks. Glad to hear it.