I'd like to know which method is more accurate and conceptually correct to integrate WaterLOGSY experiments, either by integration of area or by integration based on lineshape (relative to another experiment, with different concentration of ligand for instance). I have experience in STD-NMR and in that case, as you have off and on spectrum run under the exact same conditions, the integration based on the lineshape of the signals is more convenient. In the case of WaterLOGSY, as the comparison are made between different samples and there is no on and off spectrum, I understand that doing area integration is more correct. Any comments on this?
In general, integration by fitting Gauss–Lorenz curve give a lower systematic error. For qNMR applications reference compound in NMR buffer with know concentration (DSS, TMS) is used for companion of different samples.
The correct way to quantify integrals in WaterLOGSY (WL) is relative to a measurement without the receptor, ideally with fittings. This is not so easy to do because of differential line broadening expected on your ligand in presence of its macromolecular receptor. If you make the difference WL(+receptor) - WL(-receptor) you will remove the solvating effect and you will be able to obtain a better vision of the situation. However, this is true if you have similar T1 values! This is the same problem with STD results. Hope it helps you.
Those are some really good points especially about the differential relaxation. Is it common for people to try and compensate (if so how? reference?) for that or do they just make an assumption that it's negligible and hope for the best?
- Badges
- Science method