S. Mahmoud, B. Irwin, D. Chekmarev, S. Vyas, J. Kattas, T. Whitehead, T. Mansley, J. Bikker, G. Conduit, M. Segall,
J. Comput. Aided Mol. Des., 2021, 35(11), pp. 1125-1140
DOI: 10.1007/s10822-021-00424-3

In this article, the team demonstrates the application of Alchemite™, a deep learning method which underpins our Cerella™ technology, to sensory data. 

Abstract

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. We describe the application of a state-of-the-art deep learning method, Alchemite™, to the imputation of sparse physicochemical and sensory data and compare 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 R2 between 0.26 and 0.45 over the next best method for each sensory property.

We also demonstrate that robust uncertainty estimates generated by the imputation model enable the most accurate predictions to be identified and that imputation also more accurately predicts 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.

Find out more

You can download the preprint of this paper as a PDF or view the final publication on the journal webpage. Find out more about the application of deep learning methods to sensory data in our webinar led by three authors of this paper, Samar Mahmoud and Tamsin Mansley (Optibrium) and Dmitriy Chekmarev (IFF).