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

In this work, we present a new method for predicting complex physical-chemical 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 three-dimensional reference interaction site model. We have shown that the method allows one to achieve a good accuracy of prediction of bioconcentration factor 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.

[1]  Fumio Hirata,et al.  Hydration free energy of hydrophobic solutes studied by a reference interaction site model with a repulsive bridge correction and a thermodynamic perturbation method , 2000 .

[2]  Andriy Kovalenko,et al.  An MM/3D-RISM approach for ligand binding affinities. , 2010, The journal of physical chemistry. B.

[3]  Maxim V Fedorov,et al.  Communication: Accurate hydration free energies at a wide range of temperatures from 3D-RISM. , 2015, The Journal of chemical physics.

[4]  Frank A. P. C. Gobas,et al.  A review of bioconcentration factor (BCF) and bioaccumulation factor (BAF) assessments for organic chemicals in aquatic organisms , 2006 .

[5]  Fumio Hirata,et al.  Self-consistent description of a metal–water interface by the Kohn–Sham density functional theory and the three-dimensional reference interaction site model , 1999 .

[6]  Maxim V Fedorov,et al.  Fast and General Method To Predict the Physicochemical Properties of Druglike Molecules Using the Integral Equation Theory of Molecular Liquids. , 2015, Molecular pharmaceutics.

[7]  Christopher I. Bayly,et al.  Fast, efficient generation of high‐quality atomic charges. AM1‐BCC model: II. Parameterization and validation , 2002, J. Comput. Chem..

[8]  David S. Palmer,et al.  Hydration Free Energies of Molecular Ions from Theory and Simulation. , 2016, The journal of physical chemistry. B.

[9]  M. Levesque,et al.  Solvation free-energy pressure corrections in the three dimensional reference interaction site model. , 2015, The Journal of chemical physics.

[10]  B. Montgomery Pettitt,et al.  A dielectrically consistent interaction site theory for solvent—electrolyte mixtures , 1992 .

[11]  Maxim V Fedorov,et al.  Towards a universal method for calculating hydration free energies: a 3D reference interaction site model with partial molar volume correction , 2010, Journal of physics. Condensed matter : an Institute of Physics journal.

[12]  Robert P. Sheridan,et al.  Deep Neural Nets as a Method for Quantitative Structure-Activity Relationships , 2015, J. Chem. Inf. Model..

[13]  D. Chandler,et al.  Hydrophobicity at Small and Large Length Scales , 1999 .

[14]  Paul Labute,et al.  A Cavity Corrected 3D-RISM Functional for Accurate Solvation Free Energies , 2014, Journal of chemical theory and computation.

[15]  Fumio Hirata,et al.  Potentials of mean force of simple ions in ambient aqueous solution. I. Three-dimensional reference interaction site model approach , 2000 .

[16]  M Karplus,et al.  Ion transport in a model gramicidin channel. Structure and thermodynamics. , 1991, Biophysical journal.

[17]  David S. Palmer,et al.  Solvation thermodynamics of organic molecules by the molecular integral equation theory: approaching chemical accuracy. , 2015, Chemical reviews.

[18]  David S. Palmer,et al.  Salting-out effects by pressure-corrected 3D-RISM. , 2016, The Journal of chemical physics.

[19]  Paola Gramatica,et al.  An Update of the BCF QSAR Model Based on Theoretical Molecular Descriptors , 2005 .

[20]  F. Hirata,et al.  Predicting the binding free energy of the inclusion process of 2-hydroxypropyl-β-cyclodextrin and small molecules by means of the MM/3D-RISM method , 2016, Journal of physics. Condensed matter : an Institute of Physics journal.

[21]  B. Roux,et al.  Solvation Free Energy of Polar and Nonpolar Molecules in Water: An Extended Interaction Site Integral Equation Theory in Three Dimensions , 2000 .

[22]  L. Gendre,et al.  Classical density functional theory of solvation in molecular solvents: Angular grid implementation , 2009 .

[23]  Andy Liaw,et al.  Extreme Gradient Boosting as a Method for Quantitative Structure-Activity Relationships , 2016, J. Chem. Inf. Model..

[24]  Vijay S. Pande,et al.  Molecular graph convolutions: moving beyond fingerprints , 2016, Journal of Computer-Aided Molecular Design.

[25]  Emilio Benfenati,et al.  A new hybrid system of QSAR models for predicting bioconcentration factors (BCF). , 2008, Chemosphere.

[26]  Stefan Güssregen,et al.  Thermodynamic Characterization of Hydration Sites from Integral Equation-Derived Free Energy Densities: Application to Protein Binding Sites and Ligand Series , 2017, J. Chem. Inf. Model..

[27]  Fumio Hirata,et al.  Molecular Theory of Solvation , 2004 .

[28]  Qin Tong,et al.  Molecular fingerprint-based artificial neural networks QSAR for ligand biological activity predictions. , 2012, Molecular pharmaceutics.

[29]  J. Dearden,et al.  Linear QSAR regression models for the prediction of bioconcentration factors by physicochemical properties and structural theoretical molecular descriptors. , 2007, Chemosphere.

[30]  S Dimitrov,et al.  Base-line model for identifying the bioaccumulation potential of chemicals , 2005, SAR and QSAR in environmental research.

[31]  Nobuyuki Matubayasi,et al.  Theory of solutions in the energetic representation. I. Formulation , 2000 .

[32]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[33]  Charlotte M. Deane,et al.  Freely Available Conformer Generation Methods: How Good Are They? , 2012, J. Chem. Inf. Model..

[34]  W. Goddard,et al.  UFF, a full periodic table force field for molecular mechanics and molecular dynamics simulations , 1992 .

[35]  M. Mǐsin,et al.  Can approximate integral equation theories accurately predict solvation thermodynamics , 2017, 1704.05246.

[36]  A. A. Lagunin,et al.  SAR and QSAR in Environmental Research , 2006 .

[37]  Frank A. P. C. Gobas,et al.  A Generic QSAR for Assessing the Bioaccumulation Potential of Organic Chemicals in Aquatic Food Webs , 2003 .

[38]  Carlos Simmerling,et al.  Three-dimensional molecular theory of solvation coupled with molecular dynamics in Amber. , 2010, Journal of chemical theory and computation.

[39]  Kenta Oono,et al.  Chainer : a Next-Generation Open Source Framework for Deep Learning , 2015 .

[40]  M. Levesque,et al.  Molecular Density Functional Theory of Water. , 2013, The journal of physical chemistry letters.

[41]  Igor V. Tetko,et al.  Online chemical modeling environment (OCHEM): web platform for data storage, model development and publishing of chemical information , 2011, J. Comput. Aided Mol. Des..

[42]  David Chandler,et al.  Density functional theory of nonuniform polyatomic systems. I. General formulation , 1986 .

[43]  S. Ajmani,et al.  A neural network-based QSAR approach for exploration of diverse multi-tyrosine kinase inhibitors and its comparison with a fragment- based approach. , 2013, Current computer-aided drug design.

[44]  D. Borgis,et al.  Density functional theory of solvation and its relation to implicit solvent models. , 2005, The journal of physical chemistry. B.

[45]  Benoît Roux,et al.  An Integral Equation To Describe the Solvation of Polar Molecules in Liquid Water , 1997 .