Hi Eduard,
We are not using the same type of matching we use between the Skyline document and any type of library in this case. In this case, between the iRT specification and your spectral library Skyline doesn't use the Peptide Settings - Modifications to derive light and heavy pairs that might match. Instead, Skyline just matches the modified sequence you specify in the Standard peptides list with the modified sequences it finds in your spectral library, without any extra knowledge from the Skyline settings. It is a straight string compare.

If you choose the CiRT set from the pre-defined list, you will see the modified peptide format that is required, since that is the only set with a Cysteine containing peptide:

VC[+57.0]ENIPIVLC[+57.0]GNK

In the case you are showing that would indicate that you want to use peptides ending with K[+8.0] and R[+10.0]. This won't allow you to allow for both light and heavy spectrum matches from your search engine, but presumably, you will inject enough of the heavy labeled peptides that they will be detected as reliably as we see with other iRT standards.

This should be what Skyline gives you, if you set up a targeted experiment to measure your iRT standards and then use the Calibrate button and the Use Results button to calibrate your standards, as described in the iRT tutorial.

Hope this helps. Let us know how it goes.

--Brendan

Dear Brendan,

I now continued as you suggested and I have defined the new reference RT peptides with K[+8.0] and R[+10.0] and the library calibration works. However, with the new definition, Skyline does not seem to recognize the targets as reference RT peptides and therefore, and it gives the following error (see picture). At the moment I am not sure how to circunvent this.

On a side note, I also tried using the MSRT1 kit from Sigma (isotopically-labelled peptides). They are hard-coded in Skyline, but they appear as light versions in the calculator "Standard Peptide" list.

Eduard

In the same lines, when I use my 7 custom retention time peptides, the number of required peptides is 7 according to Skyline (see screenshot). However, when I select"Add Library", Skyline fails to calculate the iRT values even though it reports to find all seven peptides and an R=1.000. It is weird, so I guess I doing something wrong.

Eduard

PS. Skyline stable version 3.7.0.10952

Hi Eduard,
I think it is more likely that you are just pushing the edges of what we have seen in use, and we just need your example files and a bit of time to think these cases through more thoroughly and improve our support to handle them better.

The iRT support code does have multiple minimum numbers in it:

BibioSpecLite (this is probably what you are running into with your library)
        private const int MIN_IRT_ALIGNMENT_POINT_COUNT = 10;

CalibrateIrtDlg
        public const int MIN_STANDARD_PEPTIDES = 5;
        public const int MIN_SUGGESTED_STANDARD_PEPTIDES = 10;

RCalcIrt
        public const int MIN_PEPTIDES_COUNT = 8;

So, I think we really need to decide what our absolute minimum is and stick to that. It looks like the current UI allows you to create something all the way down to 5 peptides. Though, I think we warn about this and make you confirm you really want to do this, if you have below 10 standards. Was this your experience?

Given that we allow you to create a 5-standard calculator (though, we advise against it :-) we should probably allow that calculator to work in all cases, which does not appear to be the case just yet.

As for all the confusion about using labeled peptides, we probably need some good examples to improve that support as well.

If your files are small, you can attach them to this support thread. Otherwise, you can post them to

http://skyline.ms/files.url

Thanks for trying these things and reporting your experience.

--Brendan