Deep reinforcement learning for de novo drug design

By Mariya Popova, Olexandr Isayev, Alexander Tropsha. 2017

1,235 citations recorded by the source corpus.

We introduce an artificial intelligence approach to de novo design of molecules with desired physical or biological properties. We have devised and implemented a novel computational strategy for de novo design of molecules with desired properties termed ReLeaSE (Reinforcement Learning for Structural Evolution). On the basis of deep and reinforcement learning (RL) approaches, ReLeaSE integrates two deep neural networks—generative and predictive—that are trained separately but are used jointly to generate novel targeted chemical libraries. ReLeaSE uses simple representation of molecules by their simplified molecular-input line-entry system (SMILES) strings only. Generative models are trained with a stack-augmented memory network to produce chemically feasible SMILES strings, and predictive models are derived to forecast the desired properties of the de novo–generated compounds. In the first phase of the method, generative and predictive models are trained separately with a supervised learning algorithm. In the second phase, both models are trained jointly with the RL approach to bias the generation of new chemical structures toward those with the desired physical and/or biological properties. In the proof-of-concept study, we have used the ReLeaSE method to design chemical libraries with a bias toward structural complexity or toward compounds with maximal, minimal, or specific range of physical properties, such as melting point or hydrophobicity, or toward compounds with inhibitory activity against Janus protein kinase 2. The approach proposed herein can find a general use for generating targeted chemical libraries of novel compounds optimized for either a single desired property or multiple properties.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Materials Chemistry

In Molecular Biology

Citations in the corpus

Learning Multimodal Graph-to-Graph Translation for Molecular Optimization (2018, neutral): “…through generative modeling (G´omez-Bombarelli et al., 2016; Segler et al., 2017; Kusner et al., 2017; Dai et al., 2018; Jin et al., 2018; Samanta et al., 2018; Li et al., 2018a) and reinforcement learning (Guimaraes et al., 2017; Olivecrona et al., 2017; Popova et al., 2018; You et al., 2018a).”
Artificial intelligence to deep learning: machine intelligence approach for drug discovery (2021, methodology): “Likewise, Popova et al. 2018 designed a deep reinforcement learning-based algorithm, referred to as ReLeaSE (https:// github. com/ isayev/ ReLea SE), for de novo drug design.”
De Novo Protein Design for Novel Folds using Guided Conditional Wasserstein Generative Adversarial Networks (gcWGAN) (2019, neutral): “,2017), RNA (Eastman et al. , 2018), small molecules (Popova et al. , 2018; Guimaraes et al. , 2017), and peptides (Muller et al. , 2018).”
Mol-CycleGAN: a generative model for molecular optimization (2019, methodology): “Also, several approaches with reinforcement learning at their cores have been used in chemical property optimization [18, 22].”
Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction (2018, background): “Tackling the challenge of organic synthesis with data-driven approaches is particularly timely as generative models in machine learning for molecules are coming of age [4, 5, 6, 7, 8, 9, 10].”
Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models (2018, background): “Much of the popularity behind SMILES in molecular generative models [40–48] has been due to the fast adaptation of sequence modeling tools such as recurrent neural networks, attention mechanisms, and dilated convolutions, among others, to the text nature of SMILES.”
A Review on Applications of Computational Methods in Drug Screening and Design (2020, background): “Artificial intelligence, e.g., machine learning and deep learning, is playing more and more important roles in the aforementioned computational methods and thus drug development [9] [10] [11] .”
Transfer Learning in Deep Reinforcement Learning: A Survey (2020, methodology): “Health Informatics : RL has been applied to various healthcare tasks [163], including automatic medical diagnosis [164], [165], health resource scheduling [166], and drug discovery and development , [167], [168], etc.”