Auto-tune and variable modification

Issue #12 resolved
Kevin Kovalchik
created an issue

Hello! I am dealing with some tmt-label peptide data. I am setting tmt labeling of the n-terminus and K as variable modifications because I would like to be able to calculate labeling efficiency, but when I do so the auto-tune always finds a low number of PSMs with 1% FDR and the optimization is aborted. This is similar to if I just left the tmt modification out entirely. But when the n-terminus tmt label is set to fixed everything goes fine.

To illustrate:

This runs fine: -at -fmods 229.1629@[

This does not: -at -vmods 229.1629@[

And obviously neither does this: -at

So I have a question regarding how the auto-tune is working. Is the first stage done with all the variable modifications turned off? This would seem to cause the problem I am seeing, when the variable modification is more on the fixed side of things.

If this is the case, is it possible to have a switch to "turn on" the n-terminus modification during the auto-tune?

Thanks, Kevin

Comments (3)

  1. Lev Levitsky repo owner

    Hello Kevin,

    the first stage does not ignore variable modifications, so this should not be happening. Overall the first stage should behave exactly like a regular search with no optimization (except maybe it will override precursor mass tolerance). It looks like something is wrong, but I have not been able to reproduce this on the first try. I'll check with some TMT data and get back to you.

  2. Log in to comment