Phosphopeptides identified from MaxQuant

Phosphopeptides identified from MaxQuant without1102  2019-01-15

Dear MacCoss Lab,

I would like to import identified phosphopeptides search from MaxQuant to Skyline (version I put modifications.xml , mqpar.xml, and msms.txt files in the same folder. When I uploaded msms.txt to build spectral library, it showed "illegal character p found in sequence xxx". I attached the error message below. Thank you.


michael plank responded:  2019-01-17
Hi Chuan-Chih,

I got the same error recently on data from MaxQuant I suspect the problem is on the MaxQuant side and was introduced with the recent version. You`ll see that there are some phosphopeptides where the phosphate is denoted as 'p' and others as ('ph'). I posted this in the MaxQuant mailing list, but haven`t heard back yet.

without1102 responded:  2019-01-17
Hi Michael,

Thank you for your response. So far, I modify the 'p' to (ph) using Excel to do further Skyline processes. Hope they can fix this problem recently.

Brendan MacLean responded:  2019-01-17
You might also be able to work around this issue by adding a copy of the "ph" modification in your modifications.xml that uses "p". That might give you an alias for the ph modification that would require less effort than modifying the msms.txt file.

Though, I suggest this with only limited knowledge of the MaxQuant output and settings formats.

Thanks for figuring out at least one workaround.

michael plank responded:  2019-01-19
Hi again,

just a word of caution when modifying the msms.txt: It appears MaxQuant actually uses pS vs S(ph) (and equivalent) so that simple change from p to (ph) is not enough. You might have noticed that already.

I like Brendan`s suggestion, but I can`t find the field in modifications.xml that stores the abbreviation used in msms.txt.

lparsons responded:  2019-02-05
I've been seeing the same problem myself. I'm pretty sure I was able to load an msms.txt file from an earlier MQ phosphopeptide search without having this problem. I posted this to the MQ forum as well, we'll see if I have any better luck getting a reply there.
I recall that a while back there was someone from the Cox group who was routinely reading the MQ forum and posting useful replies. It doesn't look like he's been around for a while, I might see if I can reach out to him directly. I also see that there is another newer version of MQ (now available, so I'm going to re-re-rerun my samples with that to see if it helps before I dive in too much deeper.