Chemically Interpretable Graph Interaction Network for Prediction of Pharmacokinetic Properties of Drug-like Molecules

Abstract

<div>Solubility of drug molecules is related to pharmacokinetic properties such as absorption and distribution, which affects the amount of drug that is available in the body for its action. Computational or experimental evaluation of solvation free energies of drug-like molecules/solute that quantify solubilities is an arduous task and hence development of reliable computationally tractable models is sought after in drug discovery tasks in pharmaceutical industry. Here, we report a novel method based on graph neural network to predict solvation free energies. Previous studies considered only the solute for solvation free energy prediction and ignored the nature of the solvent, limiting their practical applicability. The proposed model is an end-to-end framework comprising three phases namely, message passing, interaction and prediction phases. In the first phase, message passing neural network was used to compute inter-atomic interaction within both solute and solvent molecules represented as molecular graphs. In the interaction phase, features from the preceding step is used to calculate a solute-solvent interaction map, since the solvation free energy depends on how (un)favorable the solute and solvent molecules interact with each other. The calculated interaction map that captures the solute-solvent interactions along with the features from the message passing phase is used to predict the solvation free energies in the final phase. The model predicts solvation free energies involving a large number of solvents within the limits of chemical accuracy. We also show that the interaction map captures the electronic and steric factors that govern the solubility of drug-like molecules and hence is chemically interpretable.</div>

SCONES: Self-Consistent Neural Network for Protein Stability Prediction Upon Mutation

“…Such interaction maps have been previously used to chemically interpret solvation free energies of druglike molecules in organic solvents. 51,76 Here, the interaction map consists of two columns representing the reference and mutant residues, and several rows representing neighboring residues. Each cell contains the ∆G contributions predicted for the interactions between the central-residue (column) and the neighboring residue (row).…”

MEMES: Machine learning framework for Enhanced MolEcular Screening

“…Recently new methods that use modern deep/reinforcement learning have been proposed to tackle problems in molecular sciences such as physical property prediction, 8,9 drug design tasks, 10 protein structure prediction, 11–13 molecular simulations, 14–16 and de novo molecule generation. 17 Most of the deep learning models that tackle the problem of molecular generation are based on variational autoencoders, 18–21 Generative Adversarial Networks 22–24 and Reinforcement Learning.…”

Deep learning enabled inorganic material generator

“…quantum mechanical energies), conceiving of retrosynthetic pathways, drug design, protein structure prediction, where machine learning has played a crucial role. [7][8][9][10][11] Availability of large datasets in general has increased the relevance of data driven machine learning based approaches in the area of molecular sciences. [12][13][14][15][16][17][18] In the past decade, machine learning techniques have been widely used by researchers to address many challenges in the field of material science and engineering.…”