DeepGCGR: an interpretable two-layer deep learning model for the discovery of GCGR-activating compounds

Abstract

The glucagon receptor (GCGR) is a critical target for the treatment of metabolic disorders such as Type 2 Diabetes Mellitus (T2DM) and obesity. Activation of GCGR enhances systemic insulin sensitivity through paracrine stimulation of insulin secretion, presenting a promising avenue for treatment. However, the discovery of effective GCGR agonists remains a challenging and resource-intensive process, often requiring time-consuming wet-lab experiments to synthesize and screen potential compounds. Recent advances in artificial intelligence technologies have demonstrated great potential in accelerating drug discovery by streamlining screening and efficiently predicting bioactivity. In the present work, we propose DeepGCGR, a two-layer deep learning model that leverages graph convolutional networks (GCN) integrated with a multiple attention mechanism to expedite the identification of GCGR agonists. In the first layer, the model predicts the bioactivity of various compounds against GCGR, efficiently filtering large chemical libraries to identify promising candidates. In the second layer, DeepGCGR classifies high bioactive compounds based on their functional effects on GCGR signaling, identifying those with potential agonistic or antagonistic effects. Moreover, DeepGCGR was specifically applied to identify novel GCGR-regulating compounds for the treatment of T2DM from natural products derived from traditional Chinese medicine (TCM). The proposed method will not only offer an effective strategy for discovering GCGR-targeting compounds with functional activation properties but also provide new insights into the development of T2DM therapeutics.

Keywords

Artificial intelligence / Deep learning / Type 2 diabetes mellitus / G protein-coupled receptor / Natural products

References

[1] Ong KL, Stafford LK, McLaughlin SA, et al.Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet.2023; 402(10397):203-234. https://doi.org/10.1016/S0140-6736(23)01301-6.

[2] Jia YB, Liu Y, Feng LL, et al. Role of glucagon and its receptor in the pathogenesis of diabetes. Front Endocrinol. 2022;13:928016. https://doi.org/10.3389/fendo.2022.928016.

[3] Chan JCN, Lim LL, Wareham NJ, et al. The Lancet Commission on diabetes: using data to transform diabetes care and patient lives. Lancet. 2021; 396(10267):2019-2082. https://doi.org/10.1016/s0140-6736(20)32374-6.

[4] Faselis C, Katsimardou A, Imprialos K, et al. Microvascular complications of type 2 diabetes mellitus. Curr Vasc Pharmacol. 2020; 18(2):117-124. https://doi.org/10.2174/1570161117666190502103733.

[5] Lv W, Wang XQ, Xu Q, et al. Mechanisms and characteristics of sulfonylureas and glinides. Curr Top Med Chem. 2020; 20(1):37-56. https://doi.org/10.2174/1568026620666191224141617.

[6] Qaseem A, Obley AJ, Shamliyan T, et al. Newer pharmacologic treatments in adults with type 2 diabetes: a clinical guideline from the American college of physicians. Ann Intern Med. 2024; 177(5):658-666. https://doi.org/10.7326/m23-2788.

[7] Zhao RC, Lu ZG, Yang J, et al. Drug delivery system in the treatment of diabetes mellitus. Front Bioeng Biotechnol. 2020;8:880. https://doi.org/10.3389/fbioe.2020.00880.

[8] Rena G, Hardie DG, Pearson ER. The mechanisms of action of metformin. Diabetologia. 2017; 60(9):1577-1585. https://doi.org/10.1007/s00125-017-4342-z.

[9] Phua WWT, Wong MXY, Liao Z, et al. An aPPARent functional consequence in skeletal muscle physiology via peroxisome proliferator-activated receptors. Int J Mol Sci. 2018; 19(5):1425. https://doi.org/10.3390/ijms19051425.

[10] Feng JN, Jin T. Hepatic function of glucagon-like peptide-1 and its based diabetes drugs. Med Rev. 2024; 4(4): 312-325.https://doi.org/10.1515/mr-2024-0018.

[11] Troke RC, Tan TM, Bloom SR. The future role of gut hormones in the treatment of obesity. Ther Adv Chronic Dis. 2014; 5(1):4-14. https://doi.org/10.1177/2040622313506730.

[12] Coskun T, Urva S, Roell WC, et al. LY3437943, a novel triple glucagon, GIP, and GLP-1 receptor agonist for glycemic control and weight loss: from discovery to clinical proof of concept. Cell Metab. 2022; 34(9): 1234-1247. e1239.https://doi.org/10.1016/j.cmet.2022.07.013.

[13] Bossart M, Wagner M, Elvert R, et al. Effects on weight loss and glycemic control with SAR441255, a potent unimolecular peptide GLP-1/GIP/GCG receptor triagonist. Cell Metab. 2022; 34(1): 59-74. e10.https://doi.org/10.1016/j.cmet.2021.12.005.

[14] Novikoff A, Müller TD. The molecular pharmacology of glucagon agonists in diabetes and obesity. Peptides. 2023;165:171003. https://doi.org/10.1016/j.peptides.2023.171003.

[15] Sinha B, Ghosal S. Efficacy and safety of GLP-1 receptor agonists, dual agonists, and retatrutide for weight loss in adults with overweight or obesity: a bayesian NMA. Obesity. 2025.https://doi.org/10.1002/oby.24360.

[16] Sánchez-Garrido MA, Brandt SJ, Clemmensen C, et al. GLP-1/glucagon receptor co-agonism for treatment of obesity. Diabetologia. 2017; 60:1851-1861. https://doi.org/10.1007/s00125-017-4354-8.

[17] Venugopal PP, Das BK, Soorya E, et al. Effect of hydrophobic and hydrogen bonding interactions on the potency of ß-alanine analogs of G-protein coupled glucagon receptor inhibitors. Proteins. 2020; 88(2):327-344. https://doi.org/10.1002/prot.25807.

[18] Varney MJ, Benovic JL. The role of G protein-coupled receptors and receptor kinases in pancreatic β-cell function and diabetes. Pharmacol Rev. 2024; 76(2):267-299. https://doi.org/10.1124/pharmrev.123.001015.

[19] Kim T, Nason S, Antipenko J, et al. Hepatic mTORC2 signaling facilitates acute glucagon receptor enhancement of insulin-stimulated glucose homeostasis in mice. Diabetes. 2022; 71(10):2123-2135. https://doi.org/10.2337/db21-1018.

[20] Kim T, Holleman CL, Nason S, et al. Hepatic glucagon receptor signaling enhances insulin-stimulated glucose disposal in rodents. Diabetes. 2018; 67(11):2157-2166. https://doi.org/10.2337/db18-0068.

[21] Habegger KM. Cross talk between insulin and glucagon receptor signaling in the hepatocyte. Diabetes. 2022; 71(9):1842-1851. https://doi.org/10.2337/dbi22-0002.

[22] Laker RC, Egolf S, Will S, et al. GLP-1R/GCGR dual agonism dissipates hepatic steatosis to restore insulin sensitivity and rescue pancreatic β-cell function in obese male mice. Nat Commun. 2025; 16(1):4714. https://doi.org/10.1038/s41467-025-59773-4.

[23] Zhang ZH, Leng YL, Fu XX, et al. The efficacy and safety of dachaihu decoction in the treatment of type 2 diabetes mellitus: a systematic review and meta-analysis. Front Pharmacol. 2022;13:918681. https://doi.org/10.3389/fphar.2022.918681.

[24] Bi ZH, Li H, Liang YT, et al. Emerging paradigms for target discovery of traditional medicines: a genome-wide pan-GPCR perspective. The Innovation. 2025; 6(3):100774. https://doi.org/10.1016/j.xinn.2024.100774.

[25] Rodrigues T, Reker D, Schneider P, et al. Counting on natural products for drug design. Nat Chem. 2016; 8(6):531-541. https://doi.org/10.1038/nchem.2479.

[26] Berdigaliyev N, Aljofan M. An overview of drug discovery and development. Future Med Chem. 2020; 12(10):939-947. https://doi.org/10.4155/fmc-2019-0307.

[27] Chen W, Liu XS, Zhang SY, et al. Artificial intelligence for drug discovery: resources, methods, and applications. Mol Ther Nucl Acids. 2023; 31:691-702. https://doi.org/10.1016/j.omtn.2023.02.019.

[28] Chen W, Song C, Leng L, et al. The application of artificial intelligence accelerates G protein-coupled receptor ligand discovery. Engineering. 2024; 32:18-28. https://doi.org/10.1016/j.eng.2023.09.011.

[29] Mizera M, Latek D. Ligand-receptor interactions and machine learning in GCGR and GLP-1R drug discovery. Int J Mol Sci. 2021; 22(8):4060. https://doi.org/10.3390/ijms22084060.

[30] Puszkarska AM, Taddese B, Revell J, et al. Machine learning designs new GCGR/GLP-1R dual agonists with enhanced biological potency. Nat Chem. 2024; 16(9):1436-1444. https://doi.org/10.1038/s41557-024-01532-x.

[31] Pándy-Szekeres G, Caroli J, Mamyrbekov A, et al. GPCRdb in 2023: state-specific structure models using AlphaFold2 and new ligand resources. Nucleic Acids Res.2023;51(D1):D395-D402. https://doi.org/10.1093/nar/gkac1013.

[32] Nguyen T, Le H, Quinn TP, et al. GraphDTA: predicting drug-target binding affinity with graph neural networks. Bioinformatics. 2021; 37(8):1140-1147. https://doi.org/10.1093/bioinformatics/btaa921.

[33] Nair V, Hinton GE. Rectified linear units improve restricted boltzmann machines. ICML-10. 2010;807-814. https://www.cs.toronto.edu/-fritz/absps/reluICML.pdf.

[34] Tang Q, Chen W. DeepB3P: a transformer-based model for identifying blood-brain barrier penetrating peptides with data augmentation using feedback GAN. J Adv Res. 2025; 73:459-468.https://doi.org/10.1016/j.jare.2024.08.002.

[35] Trott O, Olson AJ. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem. 2010; 31(2):455-461. https://doi.org/10.1002/jcc.21334.

[36] Siu FY, He M, de Graaf C, et al. Structure of the human glucagon class B G-protein-coupled receptor. Nature. 2013; 499(7459):444-449. https://doi.org/10.1038/nature12393.

[37] Berman HM, Westbrook J, Feng Z, et al.The protein data bank. Nucleic Acids Res. 2000; 28(1):235-242. https://doi.org/10.1093/nar/28.1.235.

[38] Kim S, Chen J, Cheng TJ, et al. PubChem in 2021: new data content and improved web interfaces. Nucleic Acids Res. 2021; 49(D1):D1388-D1395. https://doi.org/10.1093/nar/gkaa971.

[39] O'Boyle NM, Banck M, James CA, et al. Open Babel: an open chemical toolbox. J Cheminf. 2011;3:33. https://doi.org/10.1186/1758-2946-3-33.

[40] Wildman SA, Crippen GM. Prediction of physicochemical parameters by atomic contributions. J Chem Inf Comput Sci. 1999; 39(5):868-873. https://doi.org/10.1021/ci990307l.

[41] Prasanna S, Doerksen RJ. Topological polar surface area: a useful descriptor in 2D-QSAR. Curr Med Chem. 2009; 16(1):21-41. https://doi.org/10.2174/092986709787002817.

[42] Lipinski CA. Drug-like properties and the causes of poor solubility and poor permeability. J Pharmacol Toxicol Methods. 2000; 44(1):235-249. https://doi.org/10.1016/s1056-8719(00)00107-6.

[43] Wang S, Che T, Levit A, et al. Structure of the D 2 dopamine receptor bound to the atypical antipsychotic drug risperidone. Nature. 2018; 555(7695):269-273. https://doi.org/10.1038/nature25758.

[44] Gerebtzoff G, Li-Blatter X, Fischer H, et al. Halogenation of drugs enhances membrane binding and permeation. Chembiochem. 2004; 5(5):676-684. https://doi.org/10.1002/cbic.200400017.

[45] Gillis EP, Eastman KJ, Hill MD, et al. Applications of fluorine in medicinal chemistry. J Med Chem. 2015; 58(21):8315-8359. https://doi.org/10.1021/acs.jmedchem.5b00258.

[46] Zhang Q, Yin WH, Chen XY, et al. F-CPI: a multimodal deep learning approach for predicting compound bioactivity changes induced by fluorine substitution. J Med Chem. 2025: 706-718.https://doi.org/10.1021/acs.jmedchem.4c02668.

[47] Shinada NK, de Brevern AG, Schmidtke P. Halogens in protein-ligand binding mechanism: a structural perspective. J Med Chem. 2019; 62(21):9341-9356. https://doi.org/10.1021/acs.jmedchem.8b01453.

[48] Lundberg SM, Lee SI. A unified approach to interpreting model predictions. NeurIPS. 2017; 30. https://proceedings.neurips.cc/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf.

[49] Xiao WK, Zhang MQ, Zhao DN, et al. TCMKD: from ancient wisdom to modern insights-A comprehensive platform for traditional Chinese medicine knowledge discovery. J Pharm Anal. 2025; 15(6):101297. https://doi.org/10.1016/j.jpha.2025.101297.

[50] Lenselink EB, Ten Dijke N, Bongers B, et al. Beyond the hype: deep neural networks outperform established methods using a ChEMBL bioactivity benchmark set. J Cheminf. 2017; 9(1):45. https://doi.org/10.1186/s13321-017-0232-0.

[51] Cao Y, Yao W, Yang T, et al. Elucidating the mechanisms of Buyang Huanwu Decoction in treating chronic cerebral ischemia: a combined approach using network pharmacology, molecular docking, and in vivo validation. Phytomedicine. 2024;132:155820. https://doi.org/10.1016/j.phymed.2024.155820.

[52] Makhoba XH, Viegas C, Mosa RA, et al. Potential impact of the multi-target drug approach in the treatment of some complex diseases. Drug Des Devel Ther. 2020; 14:3235-3249. https://doi.org/10.2147/dddt.S257494.

[53] Chen W, Yu ZY, Leng L, et al. Artificial intelligence-curated repository of gene-encoded natural diverse components from herbal medicines. The Innovation. 2025;6:101011. https://doi.org/10.1016/j.xinn.2025.101011.