Spatial Graph Attention and Curiosity-driven Policy for Antiviral Drug Discovery (2021-06-04T00:00:00.000000Z)

TL;DR

A spatial Graph Attention Network (sGAT) that leverages self-attention over both node and edge attributes as well as encoding spatial structure is presented, of considerable interest in areas such as molecular and synthetic biology and drug discovery.

Abstract

We developed Distilled Graph Attention Policy Networks (DGAPNs), a curiosity-driven reinforcement learning model to generate novel graph-structured chemical representations that optimize user-defined objectives by efficiently navigating a physically constrained domain. The framework is examined on the task of generating molecules that are designed to bind, noncovalently, to functional sites of SARS-CoV-2 proteins. We present a spatial Graph Attention Network (sGAT) that leverages self-attention over both node and edge attributes as well as encoding spatial structure -- this capability is of considerable interest in areas such as molecular and synthetic biology and drug discovery. An attentional policy network is then introduced to learn decision rules for a dynamic, fragment-based chemical environment, and state-of-the-art policy gradient techniques are employed to train the network with enhanced stability. Exploration is efficiently encouraged by incorporating innovation reward bonuses learned and proposed by random network distillation. In experiments, our framework achieved outstanding results compared to state-of-the-art algorithms, while increasing the diversity of proposed molecules and reducing the complexity of paths to chemical synthesis.

Authors

Yulun Wu

2 papers

Austin R. Clyde

3 papers

T. Brettin

5 papers

TL;DR

Abstract

Authors

References74 items

MARS: Markov Molecular Sampling for Multi-objective Drug Discovery

Efficient Exploration of Chemical Space with Docking and Deep Learning.

Comparison of structure- and ligand-based scoring functions for deep generative models: a GPCR case study

Cryo-EM structures of the SARS-CoV-2 endoribonuclease Nsp15 reveal insight into nuclease specificity and dynamics

Edge-Featured Graph Attention Network

Autonomous molecule generation using reinforcement learning and docking to develop potential novel inhibitors

Distance Encoding: Design Provably More Powerful Neural Networks for Graph Representation Learning

GPU-Accelerated Drug Discovery with Docking on the Summit Supercomputer: Porting, Optimization, and Application to COVID-19 Research

Tipiracil binds to uridine site and inhibits Nsp15 endoribonuclease NendoU from SARS-CoV-2

Faculty Opinions recommendation of Rethinking drug design in the artificial intelligence era.

Convolutional Kernel Networks for Graph-Structured Data

Crystal structure of Nsp15 endoribonuclease NendoU from SARS‐CoV‐2

CReM: chemically reasonable mutations framework for structure generation

Geom-GCN: Geometric Graph Convolutional Networks

Multi-Objective Molecule Generation using Interpretable Substructures

Rethinking drug design in the artificial intelligence era

rlpyt: A Research Code Base for Deep Reinforcement Learning in PyTorch

Accelerating AutoDock4 with GPUs and Gradient-Based Local Search.

Heterogeneous Graph Attention Network

Fast Graph Representation Learning with PyTorch Geometric

Hypergraph Convolution and Hypergraph Attention

Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models

Optimization of Molecules via Deep Reinforcement Learning

Weisfeiler and Leman Go Neural: Higher-order Graph Neural Networks

Exploration by Random Network Distillation

Molecular Hypergraph Grammar with its Application to Molecular Optimization

Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders

Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation

MolGAN: An implicit generative model for small molecular graphs

Constrained Graph Variational Autoencoders for Molecule Design

Conditional molecular design with deep generative models

Fréchet ChemNet Distance: A Metric for Generative Models for Molecules in Drug Discovery

GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models

Syntax-Directed Variational Autoencoder for Structured Data

Learning Deep Generative Models of Graphs

NeVAE: A Deep Generative Model for Molecular Graphs

Junction Tree Variational Autoencoder for Molecular Graph Generation

GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders

Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

Deep reinforcement learning for de novo drug design

Graph Attention Networks

ChemTS: an efficient python library for de novo molecular generation

Proximal Policy Optimization Algorithms

SchNet: A continuous-filter convolutional neural network for modeling quantum interactions

Attention is All you Need

Inductive Representation Learning on Large Graphs

Retrosynthetic Reaction Prediction Using Neural Sequence-to-Sequence Models

Objective-Reinforced Generative Adversarial Networks (ORGAN) for Sequence Generation Models

Molecular de-novo design through deep reinforcement learning

Neural Message Passing for Quantum Chemistry

Grammar Variational Autoencoder

Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules

Semi-Supervised Classification with Graph Convolutional Networks

Better Informed Distance Geometry: Using What We Know To Improve Conformation Generation

Trust Region Policy Optimization

Estimation of the size of drug-like chemical space based on GDB-17 data

ADADELTA: An Adaptive Learning Rate Method

http://Mcule.com: a public web service for drug discovery

Quantifying the chemical beauty of drugs.

ChEMBL: a large-scale bioactivity database for drug discovery

Principles of early drug discovery

Scoring functions and their evaluation methods for protein-ligand docking: recent advances and future directions.

Extended-Connectivity Fingerprints

AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility

Estimation of synthetic accessibility score of drug-like molecules based on molecular complexity and fragment contributions

A semiempirical free energy force field with charge‐based desolvation

SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules

Other protomers forming the homo-hexamer are shown as grey surfaces. PDB IDs 6WLC and 6WXC were used in this illustration

Spatial Graph Convolutional Networks

Convolutional Kernel Networks for Graph-Structured Data

Should at Least Be Able to Design Molecules That Dock Well

Journal of Chemical Information and Modeling

THE GENERALIZATION OF ‘STUDENT'S’ PROBLEM WHEN SEVERAL DIFFERENT POPULATION VARLANCES ARE INVOLVED

Ieee Transactions on Systems, Man and Cybernetics—part C: Applications and Reviews 1 a Survey of Actor-critic Reinforcement Learning: Standard and Natural Policy Gradients

Field of Study

Journal Information