Molecular Generation with Recurrent Neural Networks (RNNs)

By Esben Jannik Bjerrum, Richard Threlfall. 2017

126 citations recorded by the source corpus.

The potential number of drug like small molecules is estimated to be between 10^23 and 10^60 while current databases of known compounds are orders of magnitude smaller with approximately 10^8 compounds. This discrepancy has led to an interest in generating virtual libraries using hand crafted chemical rules and fragment based methods to cover a larger area of chemical space and generate chemical libraries for use in in silico drug discovery endeavors. Here it is explored to what extent a recurrent neural network with long short term memory cells can figure out sensible chemical rules and generate synthesizable molecules by being trained on existing compounds encoded as SMILES. The networks can to a high extent generate novel, but chemically sensible molecules. The properties of the molecules are tuned by training on two different datasets consisting of fragment like molecules and drug like molecules. The produced molecules and the training databases have very similar distributions of molar weight, predicted logP, number of hydrogen bond acceptors and donors, number of rotatable bonds and topological polar surface area when compared to their respective training sets. The compounds are for the most cases synthesizable as assessed with SA score and Wiley ChemPlanner.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Physical and Theoretical Chemistry

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 Physical and Theoretical Chemistry

Citations in the corpus

Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation (2018, background): “Recently, there has been signiﬁcant advances in applying deep learning models to molecule generation [15, 38, 7, 9, 22, 4, 31, 27, 34, 42].”
Learning Deep Generative Models of Graphs (2018, background): “Recently, there has been a number of papers tackling this problem by using RNN language models on SMILES string representations of molecules (Gómez-Bombarelli et al., 2016; Segler et al., 2017; Bjerrum & Threlfall, 2017).”
In silico generation of novel, drug-like chemical matter using the LSTM neural network (2017, methodology): “Bjerrum and Threlfall [7] used molecules from the ZINC database to train an LSTM network to generate novel molecules.”
Mol-CycleGAN: a generative model for molecular optimization (2019, background): “One of the main approaches is de novo design, that is using generative models to propose new molecules that are likely to possess the desired properties [3,5,15,17].”
Graph Representation Learning via Graphical Mutual Information Maximization (2020, methodology): “Recent methods towards this goal, commonly categorized as Graph Neural Networks (GNNs), have made remarkable advancements in a great many learning tasks, such as node classification [35, 42, 50], link prediction [26, 27, 49], graph alignment [12, 19, 43], molecular generation [4, 5, 46], to name a f”
3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction (2023, methodology): “The most common molecular string representation is SMILES (Weininger, 1988), where many existing language models such as RNN can be re-purposed for the molecule generation task (Bjerrum and Threlfall, 2017; G´omez-Bombarelli et al., 2018; Kus-ner et al., 2017; Segler et al., 2018).”
Generative Recurrent Networks for De Novo Drug Design (2017, unknown): “Importantly, in contrast to other related RNN models,([10,11,13]) ours does not rely on an explicit but limited SMILES vocabulary, which renders this new approach theoretically unlimited with regard to the chemical diversity of the training data.”
Generating 3D Molecular Structures Conditional on a Receptor Binding Site with Deep Generative Models (2020, methodology): “To apply RNNs, the SMILES string format [20] is used as the molecular representation [21, 22].”
A 3D Generative Model for Structure-Based Drug Design (2022, unknown): “For example, [5, 10, 26] utilize recurrent neural networks to learn a language model of SMILES strings.”