Generative deep learning enables the discovery of a potent and selective RIPK1 inhibitor

Li, Yueshan; Zhang, Liting; Wang, Yifei; Zou, Jun; Yang, Ruicheng; Luo, Xiaoman; Wu, Chunlin; Tian, Chuanshan; Xu, Haixing; Wang, Falu; Yang, Xin; Li, Linli; Yang, Shengyong

doi:10.1038/s41467-022-34692-w

Cited by 44 publications

(33 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By referring to several recently reported prospective campaigns that made full use of deep molecular generation 30, 31 , we designed the main interface named molecular generation (MG) based on an autoregressive model that took a remastered RNN as its core network. Three sub-modules that executed model pretraining, model finetuning and molecular sampling constituted the whole workflow of molecular generation, and they were implemented in three independent tabs, i.e., “MG Pretrain”, “MG Finetune” and “MG Sample”.…”

Section: Resultsmentioning

confidence: 99%

AutoMolDesigner for Antibiotic Discovery: An AI-based Open-source Software for Automated Design of Small-molecule Antibiotics

Shen,

Guo,

Han

et al. 2023

Preprint

View full text Add to dashboard Cite

Discovery of small-molecule antibiotics with novel chemotypes serves as one of the essential strategies to address antibiotic resistance. Although a considerable number of computational tools committed to molecular design have been reported, there is a deficit in the holistic and efficient tool specifically developed for small-molecule antibiotic discovery. To address this issue, we report AutoMolDesigner, a computational modeling software dedicated to small-molecule antibiotic design. It is a generalized framework comprising two functional modules, i.e., generative deep learning-enabled molecular generation and automated machine learning based-antibacterial activity/property prediction, wherein individually trained models and curated datasets are out-of-the-box for whole cell-based antibiotic screening and design. It is open-source thus allows for the incorporation of new features for flexible use. Unlike most software programs based on Linux and command lines, this application equipped with Qt-based graphical user interface can be run on personal computers with multiple operating systems, making it much easier to use for experimental scientists. The software and related materials are freely available at GitHub (https://github.com/taoshen99/AutoMolDesigner) and Zenodo (https://zenodo.org/record/8366085).

show abstract

Section: Resultsmentioning

confidence: 99%

AutoMolDesigner for Antibiotic Discovery: An AI-based Open-source Software for Automated Design of Small-molecule Antibiotics

Shen,

Guo,

Han

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…One of the fundamental objectives in drug design is to come up with small molecules that will selectively interact with the desired target. Although there are a few recent studies that present prototype models [25,26,27,28,29,30,31,32], AI-driven target-specific drug design is a highly novel and under-studied field with a great potential to contribute to rational drug design. Incorporating protein features into the process of molecule generation is the most sensible way of designing targeted molecules, which is the approach adopted in conventional structure-based drug design.…”

Section: Introductionmentioning

confidence: 99%

Target Specific De Novo Design of Drug Candidate Molecules with Graph Transformer-based Generative Adversarial Networks

Ünlü¹,

Çevrim²,

Sarıgün³

et al. 2023

Preprint

View full text Add to dashboard Cite

Discovering novel drug candidate molecules is one of the most fundamental and critical steps in drug development. Generative deep learning models, which create synthetic data given a probability distribution, have been developed with the purpose of picking completely new samples from a partially known space. Generative models offer high potential for designing de novo molecules; however, in order for them to be useful in real-life drug development pipelines, these models should be able to design target-specific molecules, which is the next step in this field. In this study, we propose a novel generative system, DrugGEN, for the de novo design of drug candidate molecules that interact with selected target proteins. The proposed system represents compounds and protein structures as graphs and processes them via serially connected two generative adversarial networks comprising graph transformers. DrugGEN is implemented with five independent models, each with a unique sample generation routine. The system

show abstract

“…From the intermediate mathematical representation of the variational graph encoder - also known as the latent space, surrogate models can then be trained to predict more complicated properties. Previous work involved utilisation of latent space include sampling in variational autoencoders to generate potent and selective RIPK1 inhibitors 13 and BRAF inhibitor development 14 .…”

Section: Introductionmentioning

confidence: 99%

Variational graph encoders: a surprisingly effective generalist algorithm for holistic computer-aided drug design

Lam

Robbe²,

Han

et al. 2023

Preprint

View full text Add to dashboard Cite

While there has been significant progress in molecular property prediction in computer-aided drug design, there is a critical need to have fast and accurate models. Many of the currently available methods are mostly specialists in predicting specific properties, leading to the use of many models side-by-side that lead to impossibly high computational overheads for the common researcher. Henceforth, the authors propose a single, generalist unified model exploiting graph convolutional variational encoders that can simultaneously predict multiple properties such as absorption, distribution, metabolism, excretion and toxicity (ADMET), target-specific docking score prediction and drug-drug interactions. Considerably, the use of this method allows for state-of-the-art virtual screening with an acceleration advantage of up to two orders of magnitude. The minimisation of a graph variational encoder's latent space also allows for accelerated development of specific drugs for targets with Pareto optimality principles considered, and has the added advantage of explainability.

show abstract

Generative deep learning enables the discovery of a potent and selective RIPK1 inhibitor

Cited by 44 publications

References 70 publications

AutoMolDesigner for Antibiotic Discovery: An AI-based Open-source Software for Automated Design of Small-molecule Antibiotics

AutoMolDesigner for Antibiotic Discovery: An AI-based Open-source Software for Automated Design of Small-molecule Antibiotics

Target Specific De Novo Design of Drug Candidate Molecules with Graph Transformer-based Generative Adversarial Networks

Variational graph encoders: a surprisingly effective generalist algorithm for holistic computer-aided drug design

Contact Info

Product

Resources

About