Bowtie2 two mismatches in seed alignment

Question

[Bohdan Khomtchouk]

Has anyone else encountered an issue with setting two mismatches on bowtie2, using the -N flag (e.g., -N 2)? The error was:

Error: -N was set to 2, but cannot be set greater than 1
Error: Encountered internal Bowtie 2 exception (#1)

I'm aware that there's a way to filter the resultant SAM file. However, is there an optimal computational way to go about returning output with two mismatches or less in the seed alignment?

P.S. Just to be clear, I'm referring to the "Alignment options" section of the Bowtie2 manual (as you can see, it doesn't explicitly forbid two or more mismatches here, it just says it makes the alignment slower.... so the question is: is this a bug in the software or is it an error in the manual?):

-N <int> Sets the number of mismatches to allowed in a seed alignment during multiseed alignment. Can be set to 0 or 1. Setting this higher makes alignment slower (often much slower) but increases sensitivity. Default: 0.

Answer 1

[Devon Ryan]

That's an error in the manual. The internals of the program won't allow more than 1 mismatch in the seed.

Edit: Actually, it's an ambiguity in the manual. It doesn't explicitly state that values can't be higher than 1, but it does imply it.

[Bohdan Khomtchouk]

Thanks for the heads-up, probably wouldn't be a bad idea to let them know. Any ideas how to allow 2 mismatches? Some papers specify they used 2 mismatches, although they may have used bowtie (not bowtie2)...

[Devon Ryan]

The only method with bowtie2 would be to make the seed size smaller and stagger it every base. Then you can just avoid needing 2 mismatches. However, you're probably correct that they used bowtie, rather than bowtie2.