More graceful handling of libraries containing nonsensical entries (e.g. water loss on molecule containing no oxygen atoms) #3107
Conversation
…xceptions/thread.view?entityId=14a60bd2-3604-103d-a666-22f53556a042&_anchor=65275 a call to ViewLibraryDlg.AddAllPeptides with an water loss adduct on a molecule with no O in it throws an exception. Instead, let's just skip it and put a warning in the immediate window. (Ideally it wouldn't be in the library in the first place, but we don't have any data on how the library was created)
…duct_molecule_mismatch
|
"PeptideTipProvider" is where we have been trying to put the code that informs the user that there is something wrong with the library entry. |
Excellent, I'll add a check for adduct applicability there. |
…duct_molecule_mismatch
public class InvalidChemicalModificationException : IOException which is thrown when a loss would remove more atoms than are found in a molecular formula e.g. water loss from C2S5H12 (no oxygen), or label would change more atoms than are in the formula e.g. 3C' on C2S5H12 We now catch this exception in LibraryExplorer and adjust display accordingly
|
|
…duct_molecule_mismatch
|
Hmm, just noticed that the bad precursor gets neither the protein nor the molecule icon. Should it get the molecule anyway, or maybe some kind of alarm icon? |
Anything that has an error is supposed to get no icon.
|
…_mismatch' of https://github.com/ProteoWizard/pwiz into Skyline/work/20240807_graceful_handling_adduct_molecule_mismatch
…duct_molecule_mismatch
…duct_molecule_mismatch
…_mismatch' of https://github.com/ProteoWizard/pwiz into Skyline/work/20240807_graceful_handling_adduct_molecule_mismatch
…duct_molecule_mismatch
* Fix a bug in m/z calculation where the mass of the adduct was not being added to molecules described only as a mass with no chemical formula. (#3151)
…k/20240807_graceful_handling_adduct_molecule_mismatch
…duct_molecule_mismatch
…duct_molecule_mismatch
|
@nickshulman can you give this the blessing? Thanks |
…duct_molecule_mismatch
…duct_molecule_mismatch
…duct_molecule_mismatch
|
Brendan had a couple of changes - show the spectra anyway, and make the explanation prominent:
|
…duct_molecule_mismatch
Show the spectrum even if m/z can't be calculated
…duct_molecule_mismatch
…_mismatch' of https://github.com/ProteoWizard/pwiz into Skyline/work/20240807_graceful_handling_adduct_molecule_mismatch
In https://skyline.ms/announcements/home/issues/exceptions/thread.view?entityId=14a60bd2-3604-103d-a666-22f53556a042&_anchor=65275 a call to ViewLibraryDlg.AddAllPeptides with an water loss adduct on a molecule with no O in it throws an exception. Instead, let's just skip it and put a warning in the immediate window. (Ideally it wouldn't be in the library in the first place, but we don't have any data on how the library was created)
Note that a water loss is tolerated for molecules described as mass only (we just deduct the mass of H2O), but when there's a chemical formula available the adduct really needs to make sense.