Machine Learning-Guided Design of Novel Janus Kinase Inhibitors for Rheumatoid Arthritis Treatment
DOI:
https://doi.org/10.54097/6gfjmn25Keywords:
Molecular design, drug design, artificial intelligence, machine learning, deep learning, rheumatoid arthritis.Abstract
Attenuation of autoimmune responses by inhibiting immune activation pathways has been a common strategy for treating autoimmune diseases. The inhibition of JAK2 protein has proven effective in a clinical setting for treating RA. However, most of the JAK inhibitors have limited isoform selectivity, resulting in broad immune suppression that elevates the risk of infection, malignancy, and cardiovascular events. These concerns promote the development of novel JAK inhibitors with improved potency and selectivity. Focusing on the JAK2 protein, we identified in-silico novel small molecules with higher binding affinity than the existing drugs through a computational screening process that combines the generation of a drug compound library, development of predicative machine learning model, and molecular dynamics-based modeling. From a generated library of 34,992 candidates, four compounds were identified with enhanced binding affinity relative to existing inhibitors. Notably, the top candidate, M-84, exhibited a binding free energy of - 47.1 kcal/mol, which is 13.4 kcal/mol lower than that of the strongest reference drug. An analysis of the interaction profiles revealed that the additional hydrogen bonding and salt-bridge interactions within the JAK2 binding pocket enabled by the additional sulfonyl and alkylammonium groups contributed to the enhanced binding affinity, suggesting that tailoring functional groups to the chemical environment of the binding pocket can guide the rational design of new inhibitors. Moreover, the computational pipeline presented here is readily transferable to the discovery of potent small-molecule inhibitors targeting other proteins.
Downloads
References
[1] D. S. Pisetsky, “Pathogenesis of autoimmune disease,” Nat. Rev. Nephrol., vol. 19, no. 8, pp. 509–524, Aug. 2023, doi: 10.1038/s41581-023-00720-1.
[2] Q. Ding et al., “Signaling pathways in rheumatoid arthritis: implications for targeted therapy,” Signal Transduct. Target. Ther., vol. 8, no. 1, p. 68, Feb. 2023, doi: 10.1038/s41392-023-01331-9.
[3] A. D’Orazio et al., “Pathogenesis of rheumatoid arthritis: one year in review 2024,” Clin. Exp. Rheumatol., Sep. 2024, doi: 10.55563/clinexprheumatol/0307ed.
[4] A. Husivargova Theofanidis, P. Mikula, V. Timkova, Z. Macejova, R. Sanderman, and I. Nagyova, “The impact of socioeconomic disparity in autoimmune diseases,” Eur. J. Public Health, vol. 35, no. Supplement_4, p. ckaf161.1084, Oct. 2025, doi: 10.1093/eurpub/ckaf161.1084.
[5] J. S. Smolen et al., “Rheumatoid arthritis,” Nat. Rev. Dis. Primer, vol. 4, no. 1, p. 18001, Feb. 2018, doi: 10.1038/nrdp.2018.1.
[6] M. Jahid, K. U. Khan, Rehan-Ul-Haq, and R. S. Ahmed, “Overview of Rheumatoid Arthritis and Scientific Understanding of the Disease,” Mediterr. J. Rheumatol., vol. 34, no. 3, p. 284, Sep. 2023, doi: 10.31138/mjr.20230801.oo.
[7] M. Ausserwinkler et al., “Rheumatoid arthritis is independently associated with metabolic Dysfunction-Associated steatotic liver disease: evidence from the paracelsus 10,000 Population-Based cohort study,” Rheumatol. Int., vol. 45, no. 10, p. 243, Oct. 2025, doi: 10.1007/s00296-025-06002-2.
[8] X. Hu, J. Li, M. Fu, X. Zhao, and W. Wang, “The JAK/STAT signaling pathway: from bench to clinic,” Signal Transduct. Target. Ther., vol. 6, no. 1, p. 402, Nov. 2021, doi: 10.1038/s41392-021-00791-1.
[9] Y. Tanaka, Y. Luo, J. J. O’Shea, and S. Nakayamada, “Janus kinase-targeting therapies in rheumatology: a mechanisms-based approach,” Nat. Rev. Rheumatol., vol. 18, no. 3, pp. 133–145, Mar. 2022, doi: 10.1038/s41584-021-00726-8.
[10] M. Bonelli et al., “Selectivity, efficacy and safety of JAKinibs: new evidence for a still evolving story,” Ann. Rheum. Dis., vol. 83, no. 2, pp. 139–160, Feb. 2024, doi: 10.1136/ard-2023-223850.
[11] B. Mikaeili, Z. A. Alqahtani, A. L. Hincapie, and J. J. Guo, “Safety of Janus kinase inhibitors in rheumatoid arthritis: a disproportionality analysis using FAERS database from 2012 to 2022,” Clin. Rheumatol., vol. 44, no. 4, pp. 1467–1474, Apr. 2025, doi: 10.1007/s10067-025-07360-9.
[12] D. A. Pappas et al., “Herpes Zoster Reactivation in Patients with Rheumatoid Arthritis: Analysis of Disease Characteristics and Disease‐Modifying Antirheumatic Drugs,” Arthritis Care Res., vol. 67, no. 12, pp. 1671–1678, Dec. 2015, doi: 10.1002/acr.22628.
[13] A. A. Patel and O. Odenike, “The Next Generation of JAK Inhibitors: An Update on Fedratinib, Momelotonib, and Pacritinib,” Curr. Hematol. Malig. Rep., vol. 15, no. 6, pp. 409–418, Dec. 2020, doi: 10.1007/s11899-020-00596-z.
[14] A. Sertkaya, T. Beleche, A. Jessup, and B. D. Sommers, “Costs of Drug Development and Research and Development Intensity in the US, 2000-2018,” JAMA Netw. Open, vol. 7, no. 6, p. e2415445, Jun. 2024, doi: 10.1001/jamanetworkopen.2024.15445.
[15] A. V. Sadybekov and V. Katritch, “Computational approaches streamlining drug discovery,” Nature, vol. 616, no. 7958, pp. 673–685, Apr. 2023, doi: 10.1038/s41586-023-05905-z.
[16] A. R. Bhat and S. Ahmed, “Artificial intelligence (AI) in drug design and discovery: A comprehensive review,” Silico Res. Biomed., vol. 1, p. 100049, 2025, doi: 10.1016/j.insi.2025.100049.
[17] F. J. N. Ferreira and A. S. Carneiro, “AI-Driven Drug Discovery: A Comprehensive Review,” ACS Omega, vol. 10, no. 23, pp. 23889–23903, Jun. 2025, doi: 10.1021/acsomega.5c00549.
[18] V. Patel and M. Shah, “Artificial intelligence and machine learning in drug discovery and development,” Intell. Med., vol. 2, no. 3, pp. 134–140, Aug. 2022, doi: 10.1016/j.imed.2021.10.001.
[19] D. B. Catacutan, J. Alexander, A. Arnold, and J. M. Stokes, “Machine learning in preclinical drug discovery,” Nat. Chem. Biol., vol. 20, no. 8, pp. 960–973, Aug. 2024, doi: 10.1038/s41589-024-01679-1.
[20] K. Atz et al., “Prospective de novo drug design with deep interactome learning,” Nat. Commun., vol. 15, no. 1, p. 3408, Apr. 2024, doi: 10.1038/s41467-024-47613-w.
[21] D. Boldini, L. Friedrich, D. Kuhn, and S. A. Sieber, “Machine Learning Assisted Hit Prioritization for High Throughput Screening in Drug Discovery,” ACS Cent. Sci., p. acscentsci.3c01517, Mar. 2024, doi: 10.1021/acscentsci.3c01517.
[22] S. Chen et al., “Deep lead optimization enveloped in protein pocket and its application in designing potent and selective ligands targeting LTK protein,” Nat. Mach. Intell., vol. 7, no. 3, pp. 448–458, Feb. 2025, doi: 10.1038/s42256-025-00997-w.
[23] K. Swanson et al., “ADMET-AI: a machine learning ADMET platform for evaluation of large-scale chemical libraries,” Bioinformatics, vol. 40, no. 7, p. btae416, Jul. 2024, doi: 10.1093/bioinformatics/btae416.
[24] M. Badaoui et al., “Combined Free-Energy Calculation and Machine Learning Methods for Understanding Ligand Unbinding Kinetics,” J. Chem. Theory Comput., vol. 18, no. 4, pp. 2543–2555, Apr. 2022, doi: 10.1021/acs.jctc.1c00924.
[25] A. Ghidini, E. Serra, and A. Cavalli, “On Free Energy Calculations in Drug Discovery,” Acc. Chem. Res., vol. 58, no. 20, pp. 3137–3145, Oct. 2025, doi: 10.1021/acs.accounts.5c00465.
[26] M. Popova, O. Isayev, and A. Tropsha, “Deep reinforcement learning for de novo drug design,” Sci. Adv., vol. 4, no. 7, p. eaap7885, Jul. 2018, doi: 10.1126/sciadv.aap7885.
[27] H. H. Loeffler et al., “Reinvent 4: Modern AI–driven generative molecule design,” J. Cheminformatics, vol. 16, no. 1, p. 20, Feb. 2024, doi: 10.1186/s13321-024-00812-5.
[28] T. D. Goddard et al., “UCSF ChimeraX: Meeting modern challenges in visualization and analysis,” Protein Sci., vol. 27, no. 1, pp. 14–25, Jan. 2018, doi: 10.1002/pro.3235.
[29] S. Zhong and X. Guan, “Count-Based Morgan Fingerprint: A More Efficient and Interpretable Molecular Representation in Developing Machine Learning-Based Predictive Regression Models for Water Contaminants’ Activities and Properties,” Environ. Sci. Technol., vol. 57, no. 46, pp. 18193–18202, Nov. 2023, doi: 10.1021/acs.est.3c02198.
[30] J. Pan, “Large language model for molecular chemistry,” Nat. Comput. Sci., vol. 3, no. 1, pp. 5–5, Jan. 2023, doi: 10.1038/s43588-023-00399-1.
[31] L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, and A. Gulin, “CatBoost: unbiased boosting with categorical features,” 2017, arXiv. doi: 10.48550/ARXIV.1706.09516.
[32] E. Heid et al., “Chemprop: A Machine Learning Package for Chemical Property Prediction,” J. Chem. Inf. Model., vol. 64, no. 1, pp. 9–17, Jan. 2024, doi: 10.1021/acs.jcim.3c01250.
[33] Z.-W. Zhao, M. Del Cueto, and A. Troisi, “Limitations of machine learning models when predicting compounds with completely new chemistries: possible improvements applied to the discovery of new non-fullerene acceptors,” Digit. Discov., vol. 1, no. 3, pp. 266–276, 2022, doi: 10.1039/D2DD00004K.
[34] L. McInnes, J. Healy, and J. Melville, “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction,” 2018, arXiv. doi: 10.48550/ARXIV.1802.03426.
[35] J. Eberhardt, D. Santos-Martins, A. F. Tillack, and S. Forli, “AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings,” J. Chem. Inf. Model., vol. 61, no. 8, pp. 3891–3898, Aug. 2021, doi: 10.1021/acs.jcim.1c00203.
[36] D. Lu, D. Luo, Y. Zhang, and B. Wang, “A Robust Induced Fit Docking Approach with the Combination of the Hybrid All-Atom/United-Atom/Coarse-Grained Model and Simulated Annealing,” J. Chem. Theory Comput., vol. 20, no. 14, pp. 6414–6423, Jul. 2024, doi: 10.1021/acs.jctc.4c00653.
[37] M. J. Abraham et al., “GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers,” SoftwareX, vol. 1–2, pp. 19–25, Sep. 2015, doi: 10.1016/j.softx.2015.06.001.
[38] C. Tian et al., “ff19SB: Amino-Acid-Specific Protein Backbone Parameters Trained against Quantum Mechanics Energy Surfaces in Solution,” J. Chem. Theory Comput., vol. 16, no. 1, pp. 528–552, Jan. 2020, doi: 10.1021/acs.jctc.9b00591.
[39] A. W. Sousa Da Silva and W. F. Vranken, “ACPYPE - AnteChamber PYthon Parser interfacE,” BMC Res. Notes, vol. 5, no. 1, p. 367, Dec. 2012, doi: 10.1186/1756-0500-5-367.
[40] J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman, and D. A. Case, “Development and testing of a general amber force field,” J. Comput. Chem., vol. 25, no. 9, pp. 1157–1174, Jul. 2004, doi: 10.1002/jcc.20035.
[41] P. Mark and L. Nilsson, “Structure and Dynamics of the TIP3P, SPC, and SPC/E Water Models at 298 K,” J. Phys. Chem. A, vol. 105, no. 43, pp. 9954–9960, Nov. 2001, doi: 10.1021/jp003020w.
[42] G. Bussi, D. Donadio, and M. Parrinello, “Canonical sampling through velocity rescaling,” J. Chem. Phys., vol. 126, no. 1, p. 014101, Jan. 2007, doi: 10.1063/1.2408420.
[43] M. Bernetti and G. Bussi, “Pressure control using stochastic cell rescaling,” J. Chem. Phys., vol. 153, no. 11, p. 114107, Sep. 2020, doi: 10.1063/5.0020514.
[44] W. F. Van Gunsteren and H. J. C. Berendsen, “A Leap-frog Algorithm for Stochastic Dynamics,” Mol. Simul., vol. 1, no. 3, pp. 173–185, Mar. 1988, doi: 10.1080/08927028808080941.
[45] M. S. Valdés-Tresanco, M. E. Valdés-Tresanco, P. A. Valiente, and E. Moreno, “gmx_MMPBSA: A New Tool to Perform End-State Free Energy Calculations with GROMACS,” J. Chem. Theory Comput., vol. 17, no. 10, pp. 6281–6291, Oct. 2021, doi: 10.1021/acs.jctc.1c00645.
[46] D. S. Spassov, M. Atanasova, and I. Doytchinova, “A role of salt bridges in mediating drug potency: A lesson from the N-myristoyltransferase inhibitors,” Front. Mol. Biosci., vol. 9, p. 1066029, Jan. 2023, doi: 10.3389/fmolb.2022.1066029.
[47] Q. Liao, “Enhanced sampling and free energy calculations for protein simulations,” in Progress in Molecular Biology and Translational Science, vol. 170, Elsevier, 2020, pp. 177–213. doi: 10.1016/bs.pmbts.2020.01.006.
[48] P. I. Frazier, “A Tutorial on Bayesian Optimization,” 2018, arXiv. doi: 10.48550/ARXIV.1807.02811.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
