No decoy peptides

support
No decoy peptides sstoychev  2019-06-29
 

Hi,

I'm trying to train an mProphet model using decoy peptides. These have been added to the document yet when trying to build the reintegration model it says "that no decay peptides are present?

 
 
Brendan MacLean responded:  2019-06-29

Hi Stoyan,
Can you send a screenshot or something? It is highly likely you are doing something wrong or missing something, given the amount of use this feature has received without ever hearing this complaint. Are you sure your decoys were added in a way that has caused them to have associated XICs and picked peaks? When you select them can you see chromatograms and peaks?

Thanks. You can also post your file, of course, (using File > Share - Complete to create a .sky.zip) to:

https://skyline.ms/files.url

And I would be happy to have a look.

--Brendan

 
sstoychev responded:  2019-06-29

Here is the message from Skyline:

System.IO.InvalidDataException: There are no decoy peptides in the current document. Uncheck the Use Decoys Box.
at pwiz.Skyline.SettingsUI.EditPeakScoringModelDlg.TrainModel(Boolean suppressWeights) in C:\proj\pwiz_x64\pwiz_tools\Skyline\SettingsUI\EditPeakScoringModelDlg.cs:line 252
at pwiz.Skyline.SettingsUI.EditPeakScoringModelDlg.TrainModelClick() in C:\proj\pwiz_x64\pwiz_tools\Skyline\SettingsUI\EditPeakScoringModelDlg.cs:line 192

 
Brendan MacLean responded:  2019-06-29

I was thinking of a screenshot showing me your targets with the Decoys node expanded to show the decoys. Maybe a decoy selected to show it has chromatograms and peaks. i.e. Prove Skyline has lost its mind and the message provides no useful information.

That is a totally normal error message in the case that you have no decoys. To better understand why you and Skyline disagree on the presence of decoys, we need to start digging into the actual presence of scorable decoy peaks, whether they are present, and if so why Skyline thinks they are not.

If you are getting this message because you added decoys after importing DIA data and so didn't actually get any decoy chromatograms and peaks to score, then I can still consider this a bug that required a clearer message, but you won't want to wait for a fix to that to continue down a path that actually works. In fact, I feel fairly certain that there is a workaround and that the quickest route to get you to what you want is to figure out how you ended up where you are currently and come up with the necessary steps to get you on the path that actually works in Skyline.

I agree, the message could be more helpful in directing you to the problem.

--Brendan

P.S. - Is this DIA data? Or some other kind of acquisition method?

 
sstoychev responded:  2019-06-29

HI Brendan,

Never mind, I just realized what I've done wrong... added the decoys after the SWATH data was already imported so indeed they duid not have associated XICs etc. Apologies, weekend brain :)

S