ZINC20—A Free Ultralarge-Scale Chemical Database for Ligand Discovery

Journal of Chemical Information and Modeling, Год журнала: 2020, Номер 60(12), С. 6065 - 6073

Опубликована: Окт. 29, 2020

Identifying and purchasing new small molecules to test in biological assays are enabling for ligand discovery, but as purchasable chemical space continues grow into the tens of billions based on inexpensive make-on-demand compounds, simply searching this becomes a major challenge. We have therefore developed ZINC20, version ZINC with two features: methods search them. As fully enumerated database, can be searched precisely using explicit atomic-level graph-based methods, such SmallWorld similarity Arthor pattern substructure search, well 3D docking. Analysis compound sets by these related tools reveals startling features. For instance, over 97% core Bemis–Murcko scaffolds libraries unavailable from "in-stock" collections. Correspondingly, number is rising almost linear fraction elaborated molecules. Thus, an 88-fold increase versus in-stock built upon 16-fold scaffolds. The library also more structurally diverse than physical libraries, massive disc- sphere-like shaped system freely available at zinc20.docking.org.

Язык: Английский

---

Computational approaches streamlining drug discovery

Nature, Год журнала: 2023, Номер 616(7958), С. 673 - 685

Опубликована: Апрель 26, 2023

Computer-aided drug discovery has been around for decades, although the past few years have seen a tectonic shift towards embracing computational technologies in both academia and pharma. This is largely defined by flood of data on ligand properties binding to therapeutic targets their 3D structures, abundant computing capacities advent on-demand virtual libraries drug-like small molecules billions. Taking full advantage these resources requires fast methods effective screening. includes structure-based screening gigascale chemical spaces, further facilitated iterative approaches. Highly synergistic are developments deep learning predictions target activities lieu receptor structure. Here we review recent advances technologies, potential reshaping whole process development, as well challenges they encounter. We also discuss how rapid identification highly diverse, potent, target-selective ligands protein can democratize process, presenting new opportunities cost-effective development safer more small-molecule treatments. Recent approaches application streamlining discussed.

Язык: Английский

---

G protein-coupled receptors: structure- and function-based drug discovery

Signal Transduction and Targeted Therapy, Год журнала: 2021, Номер 6(1)

Опубликована: Янв. 8, 2021

Abstract As one of the most successful therapeutic target families, G protein-coupled receptors (GPCRs) have experienced a transformation from random ligand screening to knowledge-driven drug design. We are eye-witnessing tremendous progresses made recently in understanding their structure–function relationships that facilitated development at an unprecedented pace. This article intends provide comprehensive overview this important field broader readership shares some common interests discovery.

Язык: Английский

---

A practical guide to large-scale docking

Nature Protocols, Год журнала: 2021, Номер 16(10), С. 4799 - 4832

Опубликована: Сен. 24, 2021

Язык: Английский

---

Structure of a Hallucinogen-Activated Gq-Coupled 5-HT2A Serotonin Receptor

Cell, Год журнала: 2020, Номер 182(6), С. 1574 - 1588.e19

Опубликована: Сен. 1, 2020

Язык: Английский

---

Deep Docking: A Deep Learning Platform for Augmentation of Structure Based Drug Discovery

ACS Central Science, Год журнала: 2020, Номер 6(6), С. 939 - 949

Опубликована: Май 19, 2020

Drug discovery is a rigorous process that requires billion dollars of investments and decades research to bring molecule "from bench bedside". While virtual docking can significantly accelerate the drug discovery, it ultimately lags current rate expansion chemical databases already exceed billions molecular records. This recent surge small molecules availability presents great opportunities, but also demands much faster screening protocols. In order address this challenge, we herein introduce Deep Docking (DD), novel deep learning platform suitable for structures in rapid, yet accurate fashion. The DD approach utilizes quantitative structure–activity relationship (QSAR) models trained on scores subsets library approximate outcome unprocessed entries and, therefore, remove unfavorable an iterative manner. use methodology conjunction with FRED program allowed rapid calculation 1.36 from ZINC15 against 12 prominent target proteins demonstrated up 100-fold data reduction 6000-fold enrichment high scoring (without notable loss favorably docked entities). protocol readily be used any was made publicly available.

Язык: Английский

---

Synthon-based ligand discovery in virtual libraries of over 11 billion compounds

Nature, Год журнала: 2021, Номер 601(7893), С. 452 - 459

Опубликована: Дек. 15, 2021

Язык: Английский

---

Circadian rhythm as a therapeutic target

Nature Reviews Drug Discovery, Год журнала: 2021, Номер 20(4), С. 287 - 307

Опубликована: Фев. 15, 2021

Язык: Английский

---

Artificial intelligence–enabled virtual screening of ultra-large chemical libraries with deep docking

Nature Protocols, Год журнала: 2022, Номер 17(3), С. 672 - 697

Опубликована: Фев. 4, 2022

Язык: Английский

---

Efficient Exploration of Chemical Space with Docking and Deep Learning

Journal of Chemical Theory and Computation, Год журнала: 2021, Номер 17(11), С. 7106 - 7119

Опубликована: Сен. 30, 2021

With the advent of make-on-demand commercial libraries, number purchasable compounds available for virtual screening and assay has grown explosively in recent years, with several libraries eclipsing one billion compounds. Today's are larger more diverse, enabling discovery more-potent hit unlocking new areas chemical space, represented by core scaffolds. Applying physics-based silico methods an exhaustive manner, where every molecule library must be enumerated evaluated independently, is increasingly cost-prohibitive. Here, we introduce a protocol machine learning-enhanced molecular docking based on active learning to dramatically increase throughput over traditional docking. We leverage novel selection that strikes balance between two objectives: (1) identifying best scoring (2) exploring large region demonstrating superior performance compared purely greedy approach. Together automated redocking top compounds, this method captures almost all high scaffolds found This applied our campaigns against D4 AMPC targets produced dozens highly potent, inhibitors, blind test MT1 target. Our recovers than 80% experimentally confirmed hits 14-fold reduction compute cost, 90% 5% model predictions, preserving diversity

Язык: Английский

---