MetaboliteFeatureDeconvolution Negative Mode Adducts

[Michael-C-Strobel]

Hello, and thank you very much for putting together and maintaining such a great resource!

I am trying to use oms.MetaboliteFeatureDeconvolution from the PyOpenMS package and am running into some issues in negative mode.

mfd = oms.MetaboliteFeatureDeconvolution()
params = mfd.getDefaults()
params.setValue("max_neutrals", 3)
if negative_mode:
    params.setValue("negative_mode", "true")
    # Working Sometimes
    params.setValue("potential_adducts", ["H-1:-:0.6", "Cl:-:0.4"])
    # Typically Not Working
    # params.setValue("potential_adducts", ["H-1:-:0.6", "Cl:-:0.4", "H-2O-1:0:0.5", "H2CO2:0:0.5"])
    # params.setValue("potential_adducts", ["H:-:0.4", "Na:-:0.25", "NH4:-:0.25", "K:-:0.1", 'H-2O-1:0:0.05'])
    # params.setValue("potential_adducts", ["H-1:-:0.6", "Cl:-:0.4", "H-1Na:0:0.05"])
params.setValue("mass_max_diff", 10.0)
params.setValue("unit", "ppm")
mfd.setParameters(params)

I have negative_mode set when all scans have polarity oms.IonSource.Polarity.NEGATIVE.

In several instances, in particular with the later attempts I receive the following error:

WARNING!!! implicit number of default adduct is negative!!! left:-2 right: -2?

Is there any chance you could give some insight into what conditions cause this error? Looks like the warning is thrown here, but I'm having some issues tracing the C++ code.

Thanks! - Michael

[jpfeuffer]

Hi!

I have not looked into the code and have never used negative mode but did you read the statement about the potential_adducts in the algorithm class? There are quite some caveats on what is supported.

https://abibuilder.cs.uni-tuebingen.de/archive/openms/Documentation/release/3.3.0/html/TOPP_MetaboliteAdductDecharger.html

[Michael-C-Strobel]

Hello, Many thanks for the reply. I had missed this portion of the statement, which I believe is what you're referring to:

Adduct element losses are written in the form 'H-2'. All provided adducts need to have the same charge sign or be neutral! Mixing of adducts with different charge directions is only allowed as neutral complexes. For example, 'H-1Na:0:0.05' can be used to model Sodium gains (with balancing deprotonation) in negative mode.

However, the provided adducts are both negative, and in the other cases negative & neutral, which I think should be okay.

The check that throws this exception is (hc_left < 0 || hc_right < 0)

Which are defined here: hc_left = (q1 - left_charges) / default_adduct.getCharge() hc_right = ...

So the bottom line is this is somehow evaluating to a negative number which I suspect is either the q1 or left_charges. Maybe charge_min , charge_max should be set to negative? I will give that a go, but certainly if you have any inclinations or debugging steps to try I would be happy to help look into this further.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.