• Corpus ID: 211146558

Reinforcement Learning for Molecular Design Guided by Quantum Mechanics

@inproceedings{Simm2020ReinforcementLF,
  title={Reinforcement Learning for Molecular Design Guided by Quantum Mechanics},
  author={Gregor N. C. Simm and Robert Pinsler and Jos{\'e} Miguel Hern{\'a}ndez-Lobato},
  booktitle={ICML},
  year={2020}
}
Automating molecular design using deep reinforcement learning (RL) holds the promise of accelerating the discovery of new chemical compounds. Existing approaches work with molecular graphs and thus ignore the location of atoms in space, which restricts them to 1) generating single organic molecules and 2) heuristic reward functions. To address this, we present a novel RL formulation for molecular design in Cartesian coordinates, thereby extending the class of molecules that can be built. Our… 
Structure-aware generation of drug-like molecules
TLDR
This work proposes a novel supervised model that generates molecular graphs jointly with 3D pose in a discretised molecular space and proposes molecules with binding scores exceeding some known ligands, which could be useful in future wet-lab studies.
Generating 3D Molecules for Target Protein Binding
TLDR
This work proposes a novel and effective framework, known as GraphBP, to generate 3D molecules that bind to given proteins by placing atoms of specific types and locations to the given binding site one by one.
Symmetry-Aware Actor-Critic for 3D Molecular Design
TLDR
A novel actor-critic architecture for 3D molecular design that can generate molecular structures unattainable with previous approaches is proposed by exploiting the symmetries of the design process through a rotationally covariant state-action representation based on a spherical harmonics series expansion.
3DLinker: An E(3) Equivariant Variational Autoencoder for Molecular Linker Design
Deep learning has achieved tremendous success in designing novel chemical compounds with de-sirable pharmaceutical properties. In this work, we focus on a new type of drug design problem— generating
A 3D Molecule Generative Model for Structure-Based Drug Design
TLDR
A 3D generative model that generates molecules given a designated 3D protein binding site that exhibits high binding affinity to specific targets and good drug properties such as drug-likeness even if the model is not explicitly optimized for them is proposed.
Artificial Intelligence in Drug Discovery: Applications and Techniques
TLDR
An overview on drug discovery and related applications are given and AI techniques are dissected into model architectures and learning paradigms, which can be reduced to two major tasks, i.e. molecular property prediction and molecule generation.
Atomistic Global Optimization X: A Python package for optimization of atomistic structures
TLDR
This paper introduces the Atomistic Global Optimization X (AGOX) framework and code, as a customizable approach to building efirst principles global optimization algorithms, and a modular way of expressing global optimizationgorithms in the freely available AGOX python package.
Bayesian Sequential Stacking Algorithm for Concurrently Designing Molecules and Synthetic Reaction Networks
TLDR
A powerful sequential Monte Carlo algorithm is presented that recursively designs a synthetic reaction network by sequentially building up single-step reactions and shows overwhelming performance in terms of computational efficiency and coverage and novelty with respect to existing compounds.
Deep learning to catalyze inverse molecular design
GEOM, energy-annotated molecular conformations for property prediction and molecular generation
TLDR
This dataset will assist benchmarking and transfer learning in two classes of tasks: inferring 3D properties from 2D molecular graphs, and developing generative models to sample 3D conformations.
...
1
2
3
4
...

References

SHOWING 1-10 OF 46 REFERENCES
Molecular de-novo design through deep reinforcement learning
TLDR
A method to tune a sequence-based generative model for molecular de novo design that through augmented episodic likelihood can learn to generate structures with certain specified desirable properties is introduced.
Optimization of parameters for semiempirical methods V: Modification of NDDO approximations and application to 70 elements
  • J. Stewart
  • Chemistry
    Journal of molecular modeling
  • 2007
TLDR
Modifications that have been made to the NDDO core-core interaction term and to the method of parameter optimization have resulted in a more complete parameter optimization, called PM6, which has, in turn, allowed 70 elements to be parameterized.
GuacaMol: Benchmarking Models for De Novo Molecular Design
TLDR
This work proposes an evaluation framework, GuacaMol, based on a suite of standardized benchmarks, to standardize the assessment of both classical and neural models for de novo molecular design, and describes a variety of single and multiobjective optimization tasks.
Proximal Policy Optimization Algorithms
We propose a new family of policy gradient methods for reinforcement learning, which alternate between sampling data through interaction with the environment, and optimizing a "surrogate" objective
A Generative Model For Electron Paths
TLDR
A method to extract approximate reaction paths from any dataset of atom-mapped reaction SMILES strings and shows that the model achieves excellent performance on an important subset of the USPTO reaction dataset, comparing favorably to the strongest baselines.
A Model to Search for Synthesizable Molecules
TLDR
This work proposes a new molecule generation model that can generate diverse, valid and unique molecules due to the useful inductive biases of modeling reactions, and allows chemists to interrogate not only the properties of the generated molecules but also the feasibility of the synthesis routes.
Atomistic structure learning
TLDR
This work introduces an atomistic structure learning algorithm (ASLA) that utilizes a convolutional neural network to build 2D compounds and layered structures atom by atom using first-principles quantum mechanical program for physical properties.
Deep learning enables rapid identification of potent DDR1 kinase inhibitors
TLDR
A machine learning model allows the identification of new small-molecule kinase inhibitors in days and is used to discover potent inhibitors of discoidin domain receptor 1 (DDR1), a kinase target implicated in fibrosis and other diseases, in 21 days.
Optimization of Molecules via
  • Deep Reinforcement Learning. Sci. Rep
  • 2019
Optimization of Molecules via Deep Reinforcement Learning
TLDR
Inspired by problems faced during medicinal chemistry lead optimization, the MolDQN model is extended with multi-objective reinforcement learning, which maximizes drug-likeness while maintaining similarity to the original molecule.
...
1
2
3
4
5
...