Can Deep Learning Blind Docking Methods be Used to Predict Allosteric Compounds? DOI Creative Commons
Eric A. Chen, Yingkai Zhang

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Allosteric compounds offer an alternative mode of inhibition to orthosteric with opportunities for selectivity and noncompetition. Structure-based drug design (SBDD) allosteric introduces complications compared their counterparts; multiple binding sites interest are considered, often is only observed in particular protein conformations. Blind docking methods show potential virtual screening ligands, deep learning methods, such as DiffDock, achieve state-of-the-art performance on protein-ligand complex prediction benchmarks traditional Vina Lin_F9. To this aim, we explore the utility a data-driven platform called minimum distance matrix representation (MDMR) retrospectively predict recently discovered inhibitors complexed Cyclin-Dependent Kinase (CDK) 2. In contrast other representations, it uses residue-residue (or residue-ligand) feature that prioritizes formation interactions. Analysis highlights variety conformations ligand modes, identify intermediate conformation heuristic-based kinase classification do not distinguish. Next, self- cross-docking assess whether can both modes if prospective success conditional selection receptor conformation, respectively. We find combined method, DiffDock followed by Lin_F9 Local Re-Docking (DiffDock + LRD), must be selected pose. summary, work value method outlines challenges SBDD compounds.

Language: Английский

Functional dynamics of G protein-coupled receptors reveal new routes for drug discovery DOI Creative Commons
Paolo Conflitti, Edward Lyman, Mark S.P. Sansom

et al.

Nature Reviews Drug Discovery, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 2, 2025

Language: Английский

Citations

4
AlphaFold accelerated discovery of psychotropic agonists targeting the trace amine–associated receptor 1 DOI Creative Commons
Alejandro Díaz‐Holguín, Marcus Saarinen,

Duc Duy Vo

et al.

Science Advances, Journal Year: 2024, Volume and Issue: 10(32)

Published: Aug. 7, 2024

Artificial intelligence is revolutionizing protein structure prediction, providing unprecedented opportunities for drug design. To assess the potential impact on ligand discovery, we compared virtual screens using structures generated by AlphaFold machine learning method and traditional homology modeling. More than 16 million compounds were docked to models of trace amine-associated receptor 1 (TAAR1), a G protein-coupled unknown target treating neuropsychiatric disorders. Sets 30 32 highly ranked from model screens, respectively, experimentally evaluated. Of these, 25 TAAR1 agonists with potencies ranging 12 0.03 μM. The screen yielded more twofold higher hit rate (60%) discovered most potent agonists. A agonist promising selectivity profile drug-like properties showed physiological antipsychotic-like effects in wild-type but not knockout mice. These results demonstrate that can accelerate discovery.

Language: Английский

Citations

12
Solute carriers: The gatekeepers of metabolism DOI
Artem Khan, Yuyang Liu, Mark Gad

et al.

Cell, Journal Year: 2025, Volume and Issue: 188(4), P. 869 - 884

Published: Feb. 1, 2025

Language: Английский

Citations

2
Docking 14 million virtual isoquinuclidines against the mu and kappa opioid receptors reveals dual antagonists-inverse agonists with reduced withdrawal effects DOI Creative Commons
Seth F. Vigneron, Shohei Ohno, João Braz

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 14, 2025

Abstract Large library docking of tangible molecules has revealed potent ligands across many targets. While make-on-demand libraries now exceed 75 billion enumerated molecules, their synthetic routes are dominated by a few reaction types, reducing diversity and inevitably leaving interesting bioactive-like chemotypes unexplored. Here, we investigate the large-scale enumeration targeted isoquinuclidines. These “natural-product-like” rare in current functionally congested, making them as receptor probes. Using modular, four-component scheme, built docked virtual over 14.6 million isoquinuclidines against both µ- κ -opioid receptors (MOR KOR, respectively). Synthesis experimental testing 18 prioritized compounds found nine with low µM affinities. Structure-based optimization low- sub- nM antagonists inverse agonists targeting receptors. Cryo-electron microscopy (cryoEM) structures illuminate origins activity on each target. In mouse behavioral studies, member series joint MOR-antagonist KOR-inverse-agonist reversed morphine-induced analgesia, phenocopying MOR-selective anti-overdose agent naloxone. Encouragingly, new molecule induced less severe opioid-induced withdrawal symptoms compared to naloxone during precipitation, did not induce conditioned-place aversion, likely reflecting reduction dysphoria due compound’s KOR-inverse agonism. The strengths weaknesses bespoke docking, for opioid polypharmacology, will be considered.

Language: Английский

Citations

1
A database for large-scale docking and experimental results DOI Creative Commons
Brendan W. Hall, Tia A. Tummino,

Khanh Tang

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 27, 2025

The rapid expansion of readily accessible compounds over the past six years has transformed molecular docking, improving hit rates and affinities. While many millions molecules may score well in a docking campaign, results are rarely fully shared, hindering benchmarking machine learning chemical space exploration methods that seek to explore expanding spaces. To address this gap, we develop website providing access recent large library campaigns, including poses, scores, vitro for campaigns against 11 targets, with 6.3 billion docked 3729 experimentally tested. In simple proof-of-concept study speaks new library's utility, use database train models predict scores find top 0.01% scoring while evaluating only 1% library. Even these studies, some interesting trends emerge: unsurprisingly, as on larger sets, they perform better; less expected, could achieve high correlations yet still fail enrich docking-discovered ligands, or even docking-ranked molecules. It will be see how more sophisticated than studies undertaken here; is openly available at lsd.docking.org.

Language: Английский

Citations

1
The structural diversity of psychedelic drug actions revealed DOI Creative Commons
Ryan H. Gumpper, Manish K. Jain, Kuglae Kim

et al.

Nature Communications, Journal Year: 2025, Volume and Issue: 16(1)

Published: March 19, 2025

There is currently a resurgence in exploring the utility of classical psychedelics to treat depression, addiction, anxiety disorders, cluster headaches, and many other neuropsychiatric disorders. A biological target these compounds, hypothesized for their therapeutic actions, 5-HT2A serotonin receptor. Here, we present 7 cryo-EM structures covering all major compound classes psychedelic non-psychedelic agonists, including β-arrestin-biased RS130-180. Identifying molecular interactions between various receptor reveals both common distinct motifs among examined chemotypes. These findings lead broader mechanistic understanding activation, which can catalyze development novel chemotypes with potential fewer side effects. The authors hallucinogenic non-hallucinogenic compounds across multiple bound receptor, shedding light onto ligand specificity signaling bias.

Language: Английский

Citations

1
TamGen: drug design with target-aware molecule generation through a chemical language model DOI Creative Commons
Kehan Wu, Yingce Xia, Pan Deng

et al.

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Oct. 29, 2024

Language: Английский

Citations

7
Allostery DOI
Mateu Montserrat‐Canals, Gabriele Cordara, Ute Krengel

et al.

Quarterly Reviews of Biophysics, Journal Year: 2025, Volume and Issue: 58

Published: Jan. 1, 2025

Abstract Allostery describes the ability of biological macromolecules to transmit signals spatially through molecule from an allosteric site – a that is distinct orthosteric binding sites primary, endogenous ligands functional or active site. This review starts with historical overview and description classical example allostery hemoglobin other well-known examples (aspartate transcarbamoylase, Lac repressor, kinases, G-protein-coupled receptors, adenosine triphosphate synthase, chaperonin). We then discuss fringe allostery, including intrinsically disordered proteins inter-enzyme influence dynamics, entropy, conformational ensembles landscapes on mechanisms, capture essence field. Thereafter, we give over central methods for investigating molecular covering experimental techniques as well simulations artificial intelligence (AI)-based methods. conclude allostery-based drug discovery, its challenges opportunities: recent advent AI-based methods, compounds are set revolutionize discovery medical treatments.

Language: Английский

Citations

1
Modeling Boltzmann-weighted structural ensembles of proteins using artificial intelligence–based methods DOI Creative Commons
Akashnathan Aranganathan, Xinyu Gu, Dedi Wang

et al.

Current Opinion in Structural Biology, Journal Year: 2025, Volume and Issue: 91, P. 103000 - 103000

Published: Feb. 8, 2025

Language: Английский

Citations

1
Empowering AlphaFold2 for protein conformation selective drug discovery with AlphaFold2-RAVE DOI Open Access
Xinyu Gu, Akashnathan Aranganathan, Pratyush Tiwary

et al.

eLife, Journal Year: 2024, Volume and Issue: 13

Published: July 1, 2024

Small molecule drug design hinges on obtaining co-crystallized ligand-protein structures. Despite AlphaFold2's strides in protein native structure prediction, its focus apo structures overlooks ligands and associated holo Moreover, designing selective drugs often benefits from the targeting of diverse metastable conformations. Therefore, direct application AlphaFold2 models virtual screening discovery remains tentative. Here, we demonstrate an based framework combined with all-atom enhanced sampling molecular dynamics induced fit docking, named AF2RAVE-Glide, to conduct computational model small binding kinase conformations, initiated sequences. We AF2RAVE-Glide workflow three different kinases their type I II inhibitors, special emphasis known inhibitors which target classical DFG-out state. These states are not easy sample AlphaFold2. Here how AF2RAVE these conformations can be sampled for high enough accuracy enable subsequent docking more than 50% success rates across calculations. believe protocol should deployable other proteins generally.

Language: Английский

Citations

5