Feature request: Monitoring [M-1] ion for small molecules

support
Feature request: Monitoring [M-1] ion for small molecules Chris Ashwood  2018-08-08
 

Hi Skyline Team,

I'm working in the small molecule space and have noticed that sometimes, when using a wide extraction window (30ppm), my "new" targets are off-by-one errors of existing targets (e.g the monoisotopic peak for one target is within 10ppm of the first C13 isotope of another small molecule). This occurs due to my library-building DDA run performing MS2 on the 1st C13 isotope, and despite attempts at correcting this using Proteowizard's precursorRefine filter, there are still some uncorrected precursor masses that could mistaken for real targets.

In this case, the ability to integrate or include the [M-1] ion in peak integration would provide a way to quickly evaluate the off-by-one error with the peak area window in Skyline, rather than inspecting the target in full-scan view for each suspect peak. I've observed that idotp with a count of 3 is not a sufficient parameter for mitigating the off-by-one error (0.91 dotp value for both the target and off-by-one target, attached picture). I don't know if anyone else would value this feature, as this issue could be significantly reduced at the hardware/method level by running at higher-resolution or centroiding the data prior to analysis with Skyline but thought I would suggest it, in case it was seen to be a useful feature.

Cheers,
Chris

 
 
Brendan MacLean responded:  2018-08-08

Hi Chris,
You may be in luck. It should always be possible to choose the "precursor [M-1]" transition for any precursor you are extracting from MS1. Just hover the mouse cursor above the precursor element in the Targets view, then click on the drop-arrow that appears to the right of the element, then click the button with the funnel icon (Filter) on the right. In the unfiltered list you should see the "precursor [M-1]" ion offered.

Using this "transition" pick-list is covered in the MS1 Filtering tutorial on page 24, where a screenshot of the unfiltered view even contains the "precursor [M-1]" ion:

https://skyline.ms/_webdav/home/software/Skyline/@files/tutorials/MS1Filtering-2_5.pdf#page=24

If you have a lot of targets you want to do this for, I guess it could get pretty tedious doing this for all of them. I can't think of a way to add this ion automatically for everything. However, the good news is, you don't need to reimport your data, because we actually always extract MS1 chromatograms for the entire precursor isotope envelope (everything accounting for at least 1% of the distribution, plus [M-1[).

Hope this works for you. Thanks for sharing your feedback.

--Brendan

 
Chris Ashwood responded:  2018-08-08

Hi Brendan,

Awesome, thanks for teaching me something new! That does exactly what I was after. Tediousness is fine, I wouldn't want to accidentally include the [M-1] in my quantitation.

Cheers,
Chris

 
Brendan MacLean responded:  2018-08-08

Good point. We also have a "Quantitative" property on transitions, which should obviously be off by default for the [M-1] ion. If it were, then there would be no danger of it being included in quant. It is far too hard to get to right now and not used by default in cases like this where it should be, but we will improve that in future releases.

Thanks again for the feedback.

--Brendan

 
roman sakson responded:  2021-02-15

Dear Skyline team,

I was recently working with something similar regarding peptides containing selenomethionine (SeMet), a rare amino acid that is often used in protein crystallography. It turned out that selenium produces quite unusual isotopic patterns, including strong M-2 signal (see attachment, if interested). Is there a hope that M-2 signals would become available for MS-1 filtering in Skyline at some point?

Greetings from Germany,

Roman

 
Brendan MacLean responded:  2021-02-15

Hi Roman,
Is your chemical formula accurately represented in Skyline so that it is accurately calculating the isotopic distribution of the peptide? If so, then Skyline should be including isotopes with >1% of the entire distribution, except for M-1, which Skyline includes even when it is supposed to be empty. This is done to check for interference or misassignment of the monoisotopic peak, i.e. when M-1 is larger than M, you probably are actually measuring a molecule which is 1 Dalton lighter than the one you are targeting.

I see what you mean about Selenium. The isotopic distribution information Skyline uses for Se is:

Mass Number = 74
Isotopic Composition = 0.89(4)
Mass Number = 76
Isotopic Composition = 9.37(29)
Mass Number = 77
Isotopic Composition = 7.63(16)
Mass Number = 78
Isotopic Composition = 23.77(28)
Mass Number = 80
Isotopic Composition = 49.61(41)
Mass Number = 82
Isotopic Composition = 8.73(22)

So, I guess its monoisotopic mass would be the value for 80Se, while 78Se would be nearly half as abundant. I guess I would expect Skyline to work all this out, if given an accurate atomic composition for the molecule.

Maybe you could give use your Skyline document for even just one of these peptides without any data imported and we could have a look at what is actually going on in the calculations Skyline is making. If you truly are not getting M-2 available, that sounds like a bug from what I would expect. In fact, I would expect you to be able to get Skyline to include M-2 automatically if you used the Transition Settings - Full-Scan setting Isotope peaks included = Percent and then a percent of the base peak, e.g. 5%. Though, in your example, it looks like 10% would be enough to include out to M-4.

Hmmm... I just did this with SecCarbamidomethyl (C) (a standard Unimod modification including Se) and it worked as I expected with the settings I just described. I got the following for MS1 precursor transitions automatically without needing to add any manually:

PEPC[+105]IDER
532.2015++
precursor [M-4] - 530.2029++ (irank 7)
precursor [M-3] - 530.7037++ (irank 6)
precursor [M-2] - 531.2025++ (irank 2)
precursor [M-1] - 531.7037++ (irank 5)
precursor - 532.2015++ (irank 1)
precursor [M+1] - 532.7031++ (irank 3)
precursor [M+2] - 533.2029++ (irank 4)
precursor [M+3] - 533.7038++ (irank 8)

So, I think if you get your settings right, Skyline is up to the task of serving this use case, without any new feature or bug fix.

Hope this helps. Looking forward to when international travel become possible again. Take care and stay well.

--Brendan

 
roman sakson responded:  2021-02-19

Hi Brendan,

thank you, this helps a lot, actually! I was not fully aware of how Skyline calculates the isotopic distribution but this makes total sense. I will test it properly and then report but thank you already!

Take care and stay well, too!

Roman