Q&A with Prof Ian Wilson - MSG Online Seminar Series - September 2020

Post date: Sep 30, 2020 7:5:15 PM

For our September blog entry we have dedicated a section to the questions that were not answered during our September 2020 Online Seminar.

Prof Ian Wilson has kindly provided the answers to the questions directed to them concerning their talk on

‘Fast complex mixture analysis with UPLC-IM-MS’

Question 1:

Do you feel that the concentration on peak number is a bit of a red herring?

Prof Ian Wilson:

I think it can be, but it clearly depends on what your trying to do. If you are trying to “Map” the metabolome (or whatever!) and determine everything in it, then obviously the more the merrier, you maximise coverage and peak/feature numbers (though you want real features and not noise!).

If on the other hand you are looking for major differences between samples from different groups that you can use to classify/differentiate between them (disease vs healthy etc). then numbers are less important than “hits” . As my mother used to say, in this case it is “Quality, not quantity that matters

--

Question 2:

That is amazing - being able to run more samples is always the goal. Do you think your predictive CCS is more accurate than predictive retention time?

Prof Ian Wilson:

I think that the predictive CCS is, and probably always will be better than Retention time prediction (warning, I have been wrong before!). As I said in the talk, if we predict CCS the results are (usually) withing 2% or better of the measured authentic standard. I would be VERY happy if LC retention time prediction for a BROAD range of chemically unrelated compounds (rather than e.g., a congeneric series) was better than 5%. But I am hopeful that further work, with more of the (bio)chemical space covered with real data, then we will get there.

But change the stationary phase and the solvent system and predictions made under one set of chromatographic conditions will probably not transfer at all well...

--

Question 3:

Do you use MS/MS fragmentation or just MS? If just MS, is there an instrument available that does MS/MS?

Prof Ian Wilson:

We use MS/MS (or MSEe). The parent ion provided by MS + some fragments obviously provides a good place to start, but the fragmentation data provided by MS/MS is really important for structure determination, and if you have a standard of the potential unknowns in the lab, then you can be rigorous and compare them for both retention and mass spectral properties (and IM if you have that option).

If you don’t have a standard (and good library spectra for your proposed unknowns don’t exist) , then obviously things are a bit more problematic. But yes – go for a mass spectrometer with MS/MS capability.