Imputation of sensory properties using deep learning 

Predicting the sensory properties of compounds is challenging due to the subjective nature of the experimental measurements. This testing relies on a panel of human participants and is therefore also expensive and time-consuming. Scientists from Intellegens, Optibrium, and IFF applied the state-of-the-art deep learning method, Alchemite™, to the imputation of sparse physicochemical and sensory data, comparing the results with conventional quantitative structure-activity relationship methods and a multi-target graph convolutional neural network.

The imputation model achieved a substantially higher accuracy of prediction, with improvements in the R^2 measure of accuracy of between 0.26 and 0.45 over the next best method for each sensory property. Robust uncertainty estimates generated by the imputation model enabled the most accurate predictions to be identified. Imputation also more accurately predicted activity cliffs, where small changes in compound structure result in large changes in sensory properties. In combination, these results demonstrate that the use of imputation, based on data from less expensive, early experiments, enables better selection of compounds for more costly studies, saving experimental time and resources.