Peak boundaries for small molecule shotgun lipidomics

support
Peak boundaries for small molecule shotgun lipidomics bharat  2017-07-19
 
Hello:

I am interested in using skyline instead of Analyst for processing data from shotgun lipidomics. These runs are generated without a column, so all we're interested is an area under the curve from a start time to an end time. After these raw areas are generated, all other work on deconvolution, quantitation, etc are done externally. So, all we would need is for the integration to be done somewhat robotically from start time t1 to end time t2. I've been very successful, thanks to the good documentation, in getting all the transitions in (>140) and the integrations done. But, I can' get to the final step, which is this automatic peak boundary picking. So, you can see from the attachment that sometimes we get the good, and sometimes the needs adjustment. I've tried the "import peak boundaries" option, but that doesn't seem to work for me for small molecules. Anything I'm missing here?

Thanks

Bharat
 
 
Brian Pratt responded:  2017-07-19
Hi Bharat,

It's possible that the Import Peak Boundaries feature hasn't been adapted to small molecules yet - may I see your Skyline document (use the Skyline "File | Share" menu for that, please), the raw file, and the file you're using with Import Peak Boundaries? Hopefully we can get something working for you - I'll contact you privately about sending the files.

Thanks for using the Skyline support board!

Brian Pratt
 
Brian Pratt responded:  2017-07-21
Just leaving a note here for anyone that might be interested: we determined that this works fine for small molecules, but it's important to set up your transition list such that the "peptides" (molecules) names match the boundary file you're importing. In this case the "proteins" (molecule groups) were getting the name instead.
 
m j noga responded:  2021-02-14
Hi Brian,
I ended up here searching for a solution for a related problem. Would it be possible for you to give more details on how you managed to make the import of peak boundaries work for small molecules?
From what I see the molecule identifier in boundary file has to be 'PeptideModifiedSequence' and when looking at your answer above there should be a way that I can fill 'PeptideModifiedSequence' (or its synonym) for my small molecules somehow. Right now in my transition list I use 'Precursor Name' and so far I also tried 'Molecule' or 'Compound', but so far nothing seems to work. Adding 'PeptideModifiedSequence' in transition list triggers error on import that there are no valid precursors.
I will appreciate any help you could provide.
Marek
 
Nick Shulman responded:  2021-02-14
Marek,

When importing peak boundaries, even for small molecules, the molecule name should appear in a column called "PeptideModifiedSequence".

Another thing to note that is that if there is a error, Skyline might give you an error message talking about the column "MoleculeModifiedSequence". That error message is highly misleading because of a bug that we need to fix. Skyline never looks for a column called "MoleculeModifiedSequence". Even though that error message says that it needs a column called "MoleculeModifiedSequence", the column in your CSV file needs to be called "PeptideModifiedSequence".

The other columns to include in your CSV file are "Min Start Time", "Max End Time", and "File Name"

I hope this helps. If this still isn't working for you, you should send us your Skyline document and CSV file.

In Skyline you can use the menu item:
File > Share
to create a .zip file containing your Skyline document and supporting files including extracted chromatograms.

If that .zip file is less than 50MB then you can attach it to this support request. Otherwise you can upload it here:
https://skyline.ms/files.url

Remember to send us your CSV file in addition to that .sky.zip.
-- Nick
 
m j noga responded:  2021-02-14
Dear Nick,
Thanks for your quick response!
Indeed, "MoleculeModifiedSequence" is misleading but I managed to get through that and I'm using peak boundary file with "PeptideModifiedSequence" instead. I think my problem is that Skyline doesn't map these "PeptideModifiedSequence"s to my molecule names. I provide these as 'Precursor Name' in transition list .csv.
Also, I cannot re-import peak boundaries exported with build-in peak boundary report - this one is exported with blank "PeptideModifiedSequence" column.
Please see a minimal example attached, with skyline document, transition list, peak boundary file and error message.
I would very much appreciate if you could point out what I'm missing.
Marek
 
Nick Shulman responded:  2021-02-14
Marek,

The "Charge" column is optional in your peak boundary input file. You can remove it, and your peak boundaries will import successfully (like in the attached .csv file).

If you do want to include the Charge column in your small molecule peak boundaries file, what you type in there needs to match the whole adduct text for the molecules, such as "[M-H]".
Unfortunately, this does not actually work for your molecules with a positive charge. Even if you put "[M+H]" in there, it will still not match with the molecules that it is supposed to. This is another bug that we should fix. (The bug has something to do with Skyline being confused about whether that is a proteomic or a small molecule adduct).

I think the best thing for you to do is remove the "Charge" column from your CSV file.
-- Nick
 
m j noga responded:  2021-02-16
Dear Nick,
Thank you! It indeed works well without charge column, also when I scale it up.
I actually don't really need charge-specific peak boundaries, I inherited the column from the build-in Peak Boundaries report.
I can imagine it might be tricky to solve these bugs soon but maybe you could consider adding "Small Molecule Peak Boundaries" report allowing round-trip export-import of peak boundaries for small molecules? I can imagine this could maybe help others to navigate around these bugs.
Best regards,
Marek