Chromatograms not imported / visualized for small molecules fredrik edfors  2018-12-08 15:08
 

Hi,

Thanks for the awesome software. I usually work with proteomics but got interested in the small molecules tab and I'm exploring some of our metabolomics datasets using Skyline. I'm getting an inconsistent import of chromatograms, and it seems like skyline has imported everything but cannot visualize a random subset of the identified peaks. It will return an area but says that the Chromatogram is unavailable).

Please see the minimal dataset enclosed. I can see 9-Decenoylcarnitine for example, but not a single tracer from Lactic for instance (even thus I get a bar in the peak are graph). It seems like this error also occurs in the middle of a run for a set of my tracers when working with an extensive list of compounds (500+) but could be solved by dividing the set of molecules into a smaller amount. This set is however only covering 21 precursors so I don't know how to explain it.

Cheers, and grateful for any help.

Fredrik
 
 
fredrik edfors responded:  2018-12-08 15:53

Hi,

After investigating which peaks that work and which that won't work I came to the following conclusion: It seems like all my molecules including CAS and HMDB ids disappear. I tried excluding these columns from my import screen and it works like a charm. Hope that this will help you identify what's causing the issue.

Best,

Fredrik
 
Nick Shulman responded:  2018-12-08 16:08
It looks like Skyline has a bug in terms of handling molecules that have both a CAS number and a HMDB number.

This problem only manifests itself if you have extracted chromatograms from more than one .raw file. In the code that we have for merging two .skyd files, Skyline is incorrectly adding an extra backslash between the CAS number and HMDB number, and so Skyline is never able to find the chromatograms that it is looking for.

Thanks for letting us know about this bug. We will try to fix this very soon.

This problem will happen if the molecule has more than one of the things that we consider "other keys", which are CAS, HMDB, InChi, and Smiles. It looks like "InChiKey" is okay, since it gets stored in a slightly different location.
-- Nick
 
fredrik edfors responded:  2018-12-08 16:12

Hi again,

Seems like the problem isn't fully solved. This picture exemplifies one issue where the chromatogram has been partially extracted from the raw-file. The first sample has been imported between 14.15-14.65 minutes, 2nd sample between 12 and 17 minutes and the other has been imported across the whole gradient. All of them have been imported at the same time, with "include all matching scans" selected. I can see the other non-imported transitions using Xcalibur.
 
Nick Shulman responded:  2018-12-08 16:27
This looks like a different problem.

When you have both MS1 and MS2 transitions for your molecule, Skyline truncates the length of all of the chromatograms down to just the length of the MS2 chromatograms. The reason that Skyline does this is because it's the right thing to do for scheduled PRM data.
I cannot tell from your screenshot which molecule you are looking at, and whether it has any MS2 chromatograms that might have triggered this behavior.

If you think something else is going on, you should send us your new .sky.zip file and .raw file.
If those files are less than 50MB then you can attach them to this support request. Otherwise, you can upload them here:
https://skyline.ms/files.url
-- Nick