Use of C-13 NMR spectrometric data to produce a predictive model of estrogen receptor binding activity

We have developed a spectroscopic data-activity relationship (SDAR) model b ased on C-13 NMR spectral data for 30 estrogenic chemicals whose relative b inding affinities (RBA) are available for the alpha (ER alpha) and beta (ER P) estrogen receptors. The SDAR models segregated the 30 compounds into str ong and medium binding affinities. The SDAR model gave a leave-one-out (LOO ) cross-validation of 90%. Two compounds that were classified incorrectly i n the SDAR model were in the transition zone between classifications. Real and predicted C-13 NMR chemical shifts were used with test compounds to eva luate the predictive behavior of the SDAR model. The C-13 NMR SDAR model us ing predicted C-13 NMR data for the test compounds provides a rapid, reliab le, and simple way to screen whether a compound binds to the estrogen recep tors.