Autonomous Molecule Generation Using Reinforcement Learning and Docking to Develop Potential Novel Inhibitors

Overview

Journal Sci Rep

Specialty Science

Date 2020 Dec 17

PMID 33328504

Citations 17

Authors

Woosung Jeon

Dongsup Kim

Affiliations

Soon will be listed here.

Abstract

We developed a computational method named Molecule Optimization by Reinforcement Learning and Docking (MORLD) that automatically generates and optimizes lead compounds by combining reinforcement learning and docking to develop predicted novel inhibitors. This model requires only a target protein structure and directly modifies ligand structures to obtain higher predicted binding affinity for the target protein without any other training data. Using MORLD, we were able to generate potential novel inhibitors against discoidin domain receptor 1 kinase (DDR1) in less than 2 days on a moderate computer. We also demonstrated MORLD's ability to generate predicted novel agonists for the D dopamine receptor (D4DR) from scratch without virtual screening on an ultra large compound library. The free web server is available at http://morld.kaist.ac.kr .

Citing Articles

Multiscale drug screening for cardiac fibrosis identifies MD2 as a therapeutic target.

Zhang H, Thai P, Shivnaraine R, Ren L, Wu X, Siepe D Cell. 2024; 187(25):7143-7163.e22.

PMID: 39413786 PMC: 11645214. DOI: 10.1016/j.cell.2024.09.034.

Structure-aware dual-target drug design through collaborative learning of pharmacophore combination and molecular simulation.

Chen S, Xie J, Ye R, Xu D, Yang Y Chem Sci. 2024; 15(27):10366-10380.

PMID: 38994407 PMC: 11234869. DOI: 10.1039/d4sc00094c.

Mining for Potent Inhibitors through Artificial Intelligence and Physics: A Unified Methodology for Ligand Based and Structure Based Drug Design.

Li J, Zhang O, Sun K, Wang Y, Guan X, Bagni D J Chem Inf Model. 2024; 64(24):9082-9097.

PMID: 38843070 PMC: 11683870. DOI: 10.1021/acs.jcim.4c00634.

Improving Molecule Generation and Drug Discovery with a Knowledge-enhanced Generative Model.

Malusare A, Aggarwal V ArXiv. 2024; .

PMID: 38410649 PMC: 10896363.

Geometric Deep Learning for Structure-Based Ligand Design.

Powers A, Yu H, Suriana P, Koodli R, Lu T, Paggi J ACS Cent Sci. 2024; 9(12):2257-2267.

PMID: 38161364 PMC: 10755842. DOI: 10.1021/acscentsci.3c00572.

References

Aoyagi-Scharber M, Gardberg A, Yip B, Wang B, Shen Y, Fitzpatrick P . Structural basis for the inhibition of poly(ADP-ribose) polymerases 1 and 2 by BMN 673, a potent inhibitor derived from dihydropyridophthalazinone. Acta Crystallogr F Struct Biol Commun. 2014; 70(Pt 9):1143-9. PMC: 4157409. DOI: 10.1107/S2053230X14015088. View

Allen W, Fochtman B, Balius T, Rizzo R . Customizable de novo design strategies for DOCK: Application to HIVgp41 and other therapeutic targets. J Comput Chem. 2017; 38(30):2641-2663. PMC: 5659719. DOI: 10.1002/jcc.25052. View

Walters W, Murcko M . Assessing the impact of generative AI on medicinal chemistry. Nat Biotechnol. 2020; 38(2):143-145. DOI: 10.1038/s41587-020-0418-2. View

Chu Y, He X . MoleGear: A Java-Based Platform for Evolutionary De Novo Molecular Design. Molecules. 2019; 24(7). PMC: 6479339. DOI: 10.3390/molecules24071444. View

Dolinsky T, Czodrowski P, Li H, Nielsen J, Jensen J, Klebe G . PDB2PQR: expanding and upgrading automated preparation of biomolecular structures for molecular simulations. Nucleic Acids Res. 2007; 35(Web Server issue):W522-5. PMC: 1933214. DOI: 10.1093/nar/gkm276. View

Lyu J, Wang S, Balius T, Singh I, Levit A, Moroz Y . Ultra-large library docking for discovering new chemotypes. Nature. 2019; 566(7743):224-229. PMC: 6383769. DOI: 10.1038/s41586-019-0917-9. View

Labbe C, Rey J, Lagorce D, Vavrusa M, Becot J, Sperandio O . MTiOpenScreen: a web server for structure-based virtual screening. Nucleic Acids Res. 2015; 43(W1):W448-54. PMC: 4489289. DOI: 10.1093/nar/gkv306. View

Lee A, Kim D . CRDS: Consensus Reverse Docking System for target fishing. Bioinformatics. 2019; 36(3):959-960. DOI: 10.1093/bioinformatics/btz656. View

Hartenfeller M, Schneider G . De novo drug design. Methods Mol Biol. 2010; 672:299-323. DOI: 10.1007/978-1-60761-839-3_12. View

10.

Velagapudi U, Langelier M, Delgado-Martin C, Diolaiti M, Bakker S, Ashworth A . Design and Synthesis of Poly(ADP-ribose) Polymerase Inhibitors: Impact of Adenosine Pocket-Binding Motif Appendage to the 3-Oxo-2,3-dihydrobenzofuran-7-carboxamide on Potency and Selectivity. J Med Chem. 2019; 62(11):5330-5357. PMC: 7378964. DOI: 10.1021/acs.jmedchem.8b01709. View

11.

Alhossary A, Handoko S, Mu Y, Kwoh C . Fast, accurate, and reliable molecular docking with QuickVina 2. Bioinformatics. 2015; 31(13):2214-6. DOI: 10.1093/bioinformatics/btv082. View

12.

Schneider P, Walters W, Plowright A, Sieroka N, Listgarten J, Goodnow Jr R . Rethinking drug design in the artificial intelligence era. Nat Rev Drug Discov. 2019; 19(5):353-364. DOI: 10.1038/s41573-019-0050-3. View

13.

Bickerton G, Paolini G, Besnard J, Muresan S, Hopkins A . Quantifying the chemical beauty of drugs. Nat Chem. 2012; 4(2):90-8. PMC: 3524573. DOI: 10.1038/nchem.1243. View

14.

Papeo G, Posteri H, Borghi D, Busel A, Caprera F, Casale E . Discovery of 2-[1-(4,4-Difluorocyclohexyl)piperidin-4-yl]-6-fluoro-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (NMS-P118): A Potent, Orally Available, and Highly Selective PARP-1 Inhibitor for Cancer Therapy. J Med Chem. 2015; 58(17):6875-98. DOI: 10.1021/acs.jmedchem.5b00680. View

15.

Zhavoronkov A, Ivanenkov Y, Aliper A, Veselov M, Aladinskiy V, Aladinskaya A . Deep learning enables rapid identification of potent DDR1 kinase inhibitors. Nat Biotechnol. 2019; 37(9):1038-1040. DOI: 10.1038/s41587-019-0224-x. View

16.

Blaschke T, Olivecrona M, Engkvist O, Bajorath J, Chen H . Application of Generative Autoencoder in De Novo Molecular Design. Mol Inform. 2017; 37(1-2). PMC: 5836887. DOI: 10.1002/minf.201700123. View

17.

Zhou Z, Kearnes S, Li L, Zare R, Riley P . Optimization of Molecules via Deep Reinforcement Learning. Sci Rep. 2019; 9(1):10752. PMC: 6656766. DOI: 10.1038/s41598-019-47148-x. View

18.

Ruiz-Carmona S, Alvarez-Garcia D, Foloppe N, Garmendia-Doval A, Juhos S, Schmidtke P . rDock: a fast, versatile and open source program for docking ligands to proteins and nucleic acids. PLoS Comput Biol. 2014; 10(4):e1003571. PMC: 3983074. DOI: 10.1371/journal.pcbi.1003571. View

19.

Dey F, Caflisch A . Fragment-based de novo ligand design by multiobjective evolutionary optimization. J Chem Inf Model. 2008; 48(3):679-90. DOI: 10.1021/ci700424b. View

20.

Patel M, Bhatt A, Steffen J, Chergui A, Murai J, Pommier Y . Discovery and structure-activity relationship of novel 2,3-dihydrobenzofuran-7-carboxamide and 2,3-dihydrobenzofuran-3(2H)-one-7-carboxamide derivatives as poly(ADP-ribose)polymerase-1 inhibitors. J Med Chem. 2014; 57(13):5579-601. PMC: 4094269. DOI: 10.1021/jm5002502. View