Boosting Protein–Ligand Binding Pose Prediction and Virtual Screening Based on Residue–Atom Distance Likelihood Potential and Graph Transformer

Journal of Medicinal Chemistry, Год журнала: 2022, Номер 65(15), С. 10691 - 10706

Опубликована: Авг. 2, 2022

The past few years have witnessed enormous progress toward applying machine learning approaches to the development of protein-ligand scoring functions. However, robust performance and wide applicability functions remain a big challenge for increasing success rate docking-based virtual screening. Herein, novel function named RTMScore was developed by introducing tailored residue-based graph representation strategy several transformer layers protein ligand representations, followed mixture density network obtain residue-atom distance likelihood potential. Our approach resolutely validated on CASF-2016 benchmark, results indicate that can outperform almost all other state-of-the-art methods in terms both docking screening powers. Further evaluation confirms robustness our not only retain its power cross-docked poses but also achieve improved as rescoring tool larger-scale

Язык: Английский

---

Structure-based drug design with geometric deep learning

Current Opinion in Structural Biology, Год журнала: 2023, Номер 79, С. 102548 - 102548

Опубликована: Фев. 25, 2023

Structure-based drug design uses three-dimensional geometric information of macromolecules, such as proteins or nucleic acids, to identify suitable ligands. Geometric deep learning, an emerging concept neural-network-based machine has been applied macromolecular structures. This review provides overview the recent applications learning in bioorganic and medicinal chemistry, highlighting its potential for structure-based discovery design. Emphasis is placed on molecular property prediction, ligand binding site pose de novo The current challenges opportunities are highlighted, a forecast future presented.

Язык: Английский

---

TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2022, Номер unknown

Опубликована: Июнь 6, 2022

Abstract Illuminating interactions between proteins and small drug molecules is a longstanding challenge in the field of discovery. Despite importance understanding these interactions, most previous works are limited by hand-designed scoring functions insufficient conformation sampling. The recently-proposed graph neural network-based methods provides alternatives to predict protein-ligand complex one-shot manner. However, neglect geometric constraints structure weaken role local functional regions. As result, they might produce unreasonable conformations for challenging targets generalize poorly novel proteins. In this paper, we propose Trigonometry-Aware Neural networKs binding prediction, TANKBind, that builds trigonometry constraint as vigorous inductive bias into model explicitly attends all possible sites each protein segmenting whole blocks. We construct contrastive losses with region negative sampling jointly optimize interaction affinity. Extensive experiments show substantial performance gains comparison state-of-the-art physics-based deep learning-based on commonly-used benchmark datasets both affinity predictions variant settings.

Язык: Английский

---

Scoring Functions for Protein-Ligand Binding Affinity Prediction Using Structure-based Deep Learning: A Review

Frontiers in Bioinformatics, Год журнала: 2022, Номер 2

Опубликована: Июнь 17, 2022

The rapid and accurate in silico prediction of protein-ligand binding free energies or affinities has the potential to transform drug discovery. In recent years, there been a growth interest deep learning methods for based on structural information complexes. These structure-based scoring functions often obtain better results than classical when applied within their applicability domain. Here we review affinity learning, focussing different types architectures, featurization strategies, data sets, training evaluation, role explainable artificial intelligence building useful models real drug-discovery applications.

Язык: Английский

---

Multimodal learning with graphs

Nature Machine Intelligence, Год журнала: 2023, Номер 5(4), С. 340 - 350

Опубликована: Апрель 3, 2023

Язык: Английский

---

Graph neural networks

Nature Reviews Methods Primers, Год журнала: 2024, Номер 4(1)

Опубликована: Март 7, 2024

Язык: Английский

---

Geometric Interaction Graph Neural Network for Predicting Protein–Ligand Binding Affinities from 3D Structures (GIGN)

The Journal of Physical Chemistry Letters, Год журнала: 2023, Номер 14(8), С. 2020 - 2033

Опубликована: Фев. 16, 2023

Predicting protein-ligand binding affinities (PLAs) is a core problem in drug discovery. Recent advances have shown great potential applying machine learning (ML) for PLA prediction. However, most of them omit the 3D structures complexes and physical interactions between proteins ligands, which are considered essential to understanding mechanism. This paper proposes geometric interaction graph neural network (GIGN) that incorporates predicting affinities. Specifically, we design heterogeneous layer unifies covalent noncovalent into message passing phase learn node representations more effectively. The also follows fundamental biological laws, including invariance translations rotations complexes, thus avoiding expensive data augmentation strategies. GIGN achieves state-of-the-art performance on three external test sets. Moreover, by visualizing learned show predictions biologically meaningful.

Язык: Английский

---

ResGen is a pocket-aware 3D molecular generation model based on parallel multiscale modelling

Nature Machine Intelligence, Год журнала: 2023, Номер 5(9), С. 1020 - 1030

Опубликована: Сен. 7, 2023

Язык: Английский

---

Artificial intelligence in drug development

Nature Medicine, Год журнала: 2025, Номер 31(1), С. 45 - 59

Опубликована: Янв. 1, 2025

Язык: Английский

---

Current strategies to address data scarcity in artificial intelligence-based drug discovery: A comprehensive review

Computers in Biology and Medicine, Год журнала: 2024, Номер 179, С. 108734 - 108734

Опубликована: Июль 3, 2024

Язык: Английский

---