I have a set of molecules with similar structure. The 1H-15N HMBC spectra of two of these molecules were taken in DMSO-d6 and for the other two they were taken in CDCl3 and one in D2O. I was unable to obtain the spectra of all four molecules in the same solvent due to solubility and other issues. I would like to see if the 15N-chemical shifts and one of the properties of the molecules correlate (They do have a strong correlation, in fact). But, is such a correlation or the subsequent regression analysis valid because all spectra were not taken in the same solvent? asked Jan 15 '18 at 10:45 Bhanu |