1 3 Theor Chem Acc (2016) 135:66 DOI 10.1007/s00214-016-1812-1 REGULAR ARTICLE Quantitative structure–activity relationships to predict sweet and non-sweet tastes Cristian Rojas 1,2 · Davide Ballabio 3 · Viviana Consonni 3 · Piercosimo Tripaldi 4 · Andrea Mauri 3 · Roberto Todeschini 3 Received: 13 October 2015 / Accepted: 18 January 2016 © Springer-Verlag Berlin Heidelberg 2016 properly validated through cross-validation and external test sets. The applicability domain of models was investigated, and the interpretation of the role of the molecular descriptors in classifying sweet and non-sweet tastes was evaluated. The classification and the performance of the models presented in this paper are simple but accurate. They are based on a relatively small number of descriptors and a straightforward classification approach. The results presented here indicate that the proposed models can be used to accurately select new compounds as potential sweetener candidates. Keywords QSAR · k-Nearest neighbors · Classification · Sweetness 1 Introduction Taste perception is a subject of study in many disciplines such as psychology, chemistry, biochemistry, pharmacol- ogy, anatomy and physiology, and especially the food industry. There appears to be an increase in research related to taste perception in recent years. The taste mechanism in humans is complex and includes apparent contradictions. For example, a chemist might assume that taste is related to chemical reasons, while a psychologist might presume that taste might be related to some brain stimulus. While both the chemist and the psychologist are, in some ways correct, studies of anatomy and physiology have shown that taste is related to soluble molecules which interact with recep- tors on the taste buds in the tongue. The sweetness, bitter- ness, saltiness and sourness are the most important primary tastes that define a substance [1]. The human perception of these primary tastes, which may vary from person to per- son, may be related to subtle differences in psychology, anatomy, or receptor function. Abstract The aim of this work was the calibration and validation of mathematical models based on a quantita- tive structure–activity relationship approach to discriminate sweet, tasteless and bitter molecules. The sweet-tasteless and the sweet-bitter datasets included 566 and 508 compounds, respectively. A total of 3763 conformation-independent Dragon molecular descriptors were calculated and subse- quently reduced through both unsupervised reduction and supervised selection coupled with the k-nearest neighbors classification technique. A model based on nine descrip- tors was retained as the optimal one for sweet and tasteless molecules, while a model based on four descriptors was calibrated for the sweetness-bitterness dataset. Models were Published as part of the special collection of articles “CHITEL 2015 - Torino - Italy”. Electronic supplementary material The online version of this article (doi:10.1007/s00214-016-1812-1) contains supplementary material, which is available to authorized users. * Cristian Rojas crojasvilla@gmail.com 1 Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas INIFTA (CCT La Plata-CONICET, UNLP), Diag. 113 y 64, C.C. 16, Sucursal 4, 1900 La Plata, Argentina 2 Decanato General de Investigaciones, Universidad del Azuay, Av. 24 de Mayo 7-77 y Hernán Malo, Apartado Postal 01.01.981, Cuenca, Ecuador 3 Milano Chemometrics and QSAR Research Group, Department of Earth and Environmental Sciences, University of Milano-Bicocca, P.za della Scienza 1, 20126 Milan, Italy 4 Laboratorio de Química-Física de Alimentos, Facultad de Ciencia y Tecnología, Universidad del Azuay, Av. 24 de Mayo 7-77 y Hernán Malo, Apartado Postal 01.01.981, Cuenca, Ecuador