I’m sorry if this is a very basic question, but I need to confirm this with somebody, who’s an expert in the field
When subtracting edit-on minus edit-off spectra in J-edited spectroscopy (say for 2-HG or GABA), do I need to scale the difference spectrum by one half when trying to quantify the metabolites using an unsuppressed water reference? In other words, treat the on/off spectra as two transients which I’m averaging?
This is an excellent question that comes up for discussion about once per year in our group meetings!
As long as you’re linear-combination modeling and applying the same operation while you generate your basis set, the scaling is intrinsically taken care of, no matter if you average the difference spectrum or not.
If you were to do some sort of peak integration (directly calculating the area under an edited peak), you would need to divide by 2. To confuse matters more, there is sometimes an additional “editing efficiency” factor that people apply when they do simple peak integration, which is usually set to 0.5 for GABA (see, for example, in Gannet)… but I guess that factor would change if you didn’t divide by 2.
(Instinctively, if we were to come up with a terminology, I guess we should indeed think of the difference spectrum as an ‘average’ across all transients, just with alternating signs, so just to avoid all confusion we probably should divide by 2… does everyone do it? Do we do it in our own software? I don’t think so!)
