3D matters! 3D-RISM and 3D convolutional neural network for accurate bioaccumulation prediction

Sosnin, Sergey and Misin, Maksim and Palmer, David S. and Fedorov, Maxim V. (2018) 3D matters! 3D-RISM and 3D convolutional neural network for accurate bioaccumulation prediction. Journal of Physics: Condensed Matter, 30. 32LT03. ISSN 0953-8984 (https://doi.org/10.1088/1361-648X/aad076)

[thumbnail of Sosnin-etal-JPCM-2018-3D-RISM-and-3D-convolutional-neural-network-for-accurate-bioaccumulation-prediction]
Preview
 Text. Filename: Sosnin_etal_JPCM_2018_3D_RISM_and_3D_convolutional_neural_network_for_accurate_bioaccumulation_prediction.pdf
Accepted Author Manuscript
Download (3MB)| Preview

Abstract

In this work, we present a new method for predicting complex physicalchemical properties of organic molecules. The approach utilizes 3D convolutional neural network (ActivNet4) that uses solvent spatial distributions around solutes as input. These spatial distributions are obtained by a molecular theory called threedimensional reference interaction site model (3D-RISM). We have shown that the method allows one to achieve a good accuracy of prediction of bioconcentration factor (BCF) which is difficult to predict by direct application of methods of molecular theory or simulations. Our research demonstrates that combination of molecular theories with modern machine learning approaches can be effectively used for predicting properties that are otherwise inaccessible to purely theory-based models.

CORE (COnnecting REpositories)