PGMG: A Pharmacophore-Guided Deep Learning Approach for Bioactive Molecular Generation

@article{Zhu2022PGMGAP,
  title={PGMG: A Pharmacophore-Guided Deep Learning Approach for Bioactive Molecular Generation},
  author={Hui Zhu and Renyi Zhou and Jing Tang and Min Li},
  journal={ArXiv},
  year={2022},
  volume={abs/2207.00821}
}
The rational design of novel molecules with desired bioactivity is a critical but challenging task in drug discovery, especially when treating a novel target family or understudied targets. Here, we propose PGMG, a pharmacophore-guided deep learning approach for bioactivate molecule generation. Through the guidance of pharmacophore, PGMG provides a flexible strategy to generate bioactive molecules with structural diversity in various scenarios using a trained variational autoencoder. We show… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 48 REFERENCES

Structure-based de novo drug design using 3D deep generative models

This work introduces DeepLigBuilder, a novel deep learning-based method for de novo drug design that generates 3D molecular structures in the binding sites of target proteins and suggests a list of drug-like compounds with novel chemical structures, high predicted affinity, and similar binding features to those of known inhibitors.

Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

This work shows that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing, and demonstrates that the properties of the generated molecules correlate very well with those of the molecules used to train the model.

Assessing the impact of generative AI on medicinal chemistry

The synthesis and testing of molecules derived from a generative model, a variation on the de novo design programs that were in vogue during the 1990s and early 2000s, and the ultimate value of generative models will be demonstrated through the synthesis and biological evaluation of the novel molecules they identify are reported.

Optimization of Molecules via Deep Reinforcement Learning

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.

Deep generative design with 3D pharmacophoric constraints

This work combines an existing graph-based deep generative model, DeLinker, with a convolutional neural network to utilise physically-meaningful 3D representations of molecules and target pharmacophores and demonstrates the benefit of doing so.

GuacaMol: Benchmarking Models for De Novo Molecular Design

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.

Protein-Ligand-Based Pharmacophores: Generation and Utility Assessment in Computational Ligand Profiling

Receptor-ligand-based pharmacophore search is notably a fast and reliable alternative to docking when few ligand information is available for some targets, and overall, the present study suggests that a workflow using the best profiling method according to the protein-ligands context is the best strategy to follow.

De novo generation of hit-like molecules from gene expression signatures using artificial intelligence

A generative model that bridges systems biology and molecular design, conditioning a generative adversarial network with transcriptomic data that can automatically design molecules that have a high probability to induce a desired transcriptomic profile.

Deep Generative Models for 3D Linker Design

This is the first molecular generative model to incorporate 3D structural information directly in the design process, and the effectiveness and applicability of this approach on a diverse range of design problems: fragment linking, scaffold hopping, and proteolysis targeting chimera (PROTAC) design.

Mol-CycleGAN: a generative model for molecular optimization

Mol-CycleGAN—a CycleGAN-based model that generates optimized compounds with high structural similarity to the original ones is introduced, and in the task of optimization of penalized logP of drug-like molecules the model significantly outperforms previous results.