Practical notes on building molecular graph generative models

Mercado, Rocío; Rastemo, Tobias; Lindelöf, Edvard; Klambauer, Günter; Engkvist, Ola; Chen, Hongming; Bjerrum, Esben Jannik

doi:10.1002/ail2.18

Cited by 22 publications

(11 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this, the DDC code available at github.com/pcko1/Deep-Drug-Coder was used. Finally, the GraphINVENT code available at github.com/MolecularAI/GraphINVENT ,, was used. All methods except for GraphINVENT are string-based generative models, whereas GraphINVENT is a graph-based generative model.…”

Section: Methodsmentioning

confidence: 99%

Comparative Study of Deep Generative Models on Chemical Space Coverage

Zhang

Mercado

Engkvist

et al. 2021

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

In recent years, deep molecular generative models have emerged as promising methods for de novo molecular design. Thanks to the rapid advance of deep learning techniques, deep learning architectures such as recurrent neural networks, variational autoencoders, and adversarial networks have been successfully employed for constructing generative models. Recently, quite a few metrics have been proposed to evaluate these deep generative models. However, many of these metrics cannot evaluate the chemical space coverage of sampled molecules. This work presents a novel and complementary metric for evaluating deep molecular generative models. The metric is based on the chemical space coverage of a reference datasetGDB-13. The performance of seven different molecular generative models was compared by calculating what fraction of the structures, ring systems, and functional groups could be reproduced from the largely unseen reference set when using only a small fraction of GDB-13 for training. The results show that the performance of the generative models studied varies significantly using the benchmark metrics introduced herein, such that the generalization capabilities of the generative models can be clearly differentiated. In addition, the coverages of GDB-13 ring systems and functional groups were compared between the models. Our study provides a useful new metric that can be used for evaluating and comparing generative models.

show abstract

Section: Methodsmentioning

confidence: 99%

Comparative Study of Deep Generative Models on Chemical Space Coverage

Zhang

Mercado

Engkvist

et al. 2021

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The most commonly used representations are SMILES strings 5 and molecular graphs. Multiple models for generating SMILES strings [6][7][8][9] and molecular graphs [10][11][12][13][14] corresponding to synthetically feasible novel molecules have been proposed. Initially, these models are typically trained on a diverse dataset of molecules so that they can generate a broad distribution of molecules.…”

mentioning

confidence: 99%

Generative and reinforcement learning approaches for the automated de novo design of bioactive compounds

et al. 2022

View full text Add to dashboard Cite

Deep generative neural networks have been used increasingly in computational chemistry for de novo design of molecules with desired properties. Many deep learning approaches employ reinforcement learning for optimizing the target properties of the generated molecules. However, the success of this approach is often hampered by the problem of sparse rewards as the majority of the generated molecules are expectedly predicted as inactives. We propose several technical innovations to address this problem and improve the balance between exploration and exploitation modes in reinforcement learning. In a proof-of-concept study, we demonstrate the application of the deep generative recurrent neural network architecture enhanced by several proposed technical tricks to design inhibitors of the epidermal growth factor (EGFR) and further experimentally validate their potency. The proposed technical solutions are expected to substantially improve the success rate of finding novel bioactive compounds for specific biological targets using generative and reinforcement learning approaches.

show abstract

“…The global readout block uses both the node-and graph-level information to predict the APD. Many different global readout block architectures were tested before selecting the one presented here, and are described elsewhere [60].…”

Section: Global Readout Blockmentioning

confidence: 99%

Graph Networks for Molecular Design

Mercado

Rastemo²,

Lindelöf³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Deep learning methods applied to chemistry can be used to accelerate the discovery of new molecules. This work introduces GraphINVENT, a platform developed for graph-based molecular design using graph neural networks (GNNs). GraphINVENT uses a tiered deep neural network architecture to probabilistically generate new molecules a single bond at a time. All models implemented in GraphINVENT can quickly learn to build molecules resembling the training set molecules without any explicit programming of chemical rules. The models have been benchmarked using the MOSES distribution-based metrics, showing how GraphINVENT models compare well with state-of-the-art generative models. This work is one of the first thorough graph-based molecular design studies, and illustrates how GNN-based models are promising tools for molecular discovery.<br>

show abstract

Practical notes on building molecular graph generative models

Cited by 22 publications

References 32 publications

Comparative Study of Deep Generative Models on Chemical Space Coverage

Comparative Study of Deep Generative Models on Chemical Space Coverage

Generative and reinforcement learning approaches for the automated de novo design of bioactive compounds

Graph Networks for Molecular Design

Contact Info

Product

Resources

About