Skip to main content
Fig. 1 | Journal of Cheminformatics

Fig. 1

From: Ilm-NMR-P31: an open-access 31P nuclear magnetic resonance database and data-driven prediction of 31P NMR shifts

Fig. 1

Different concepts how to represent molecules for NMR shift prediction. a HOSE codes look at the environment of a given atom, here P atom. b Increment models rely on the partition of the molecule into different substituents which are bound to the P atom. c Same as b but only considering atoms close to the P atom of interest. d Fingerprints look at different molecular fragments which might not have a connection to the P atom of interest

Back to article page