Exclusive

Progress of artificial intelligence-driven drug discovery

  • Xinheng HE ,
  • Sihan GAO ,
  • Junrui LI ,
  • Huaqiang XU , *
Expand
  • Shanghai Institute of Materia Medica, Chinese Academy of Science, Shanghai 201203, China

Received date: 2025-04-23

  Revised date: 2025-06-03

  Online published: 2025-07-03

Copyright

All rights reserved. Unauthorized reproduction is prohibited.

Abstract

Drug discovery, as the core driving force of the modern pharmaceutical industry, faces the difficulty of the traditional model's "high investment, long cycle, and low output, " urgently requiring breakthroughs to address increasingly complex health demands. The rapid development of artificial intelligence (AI) technology has brought revolutionary changes to drug discovery, significantly enhancing efficiency and success rates in areas such as protein structure prediction, protein design, antibody drug design, and small molecule drug development. This article focuses on the domestic and international progress of AI in these key domains, providing an in-depth analysis of AI breakthrough in protein structure prediction and its potential applications in target discovery and virtual screening. It explores the closed-loop model of AI-driven protein design, from structure prediction to functional innovation, and examines AI's role in antibody sequence optimization, affinity maturation, and novel antibody design. Additionally, it reviews the latest achievements of AI in small molecule drug target identification, virtual screening, and ADMET optimization. The article also highlights challenges in AI applications, including data quality, model interpretability, and experimental validation, while envisioning future directions such as multimodal data integration, dynamic behavior prediction, and automated platforms. By comprehensively analyzing the current state and challenges of AI-enabled drug discovery, this article aims to offer scientific perspectives and insights to accelerate new drug creation and enhance human health and well-being. It seeks to provide readers with a thorough and insightful view of technological issues in AI-empowered drug discovery, stimulate thinking about future directions, and promote the more effective application of AI technologies in this field, ultimately benefiting human health through an accelerated drug development process.

Cite this article

Xinheng HE , Sihan GAO , Junrui LI , Huaqiang XU . Progress of artificial intelligence-driven drug discovery[J]. Science & Technology Review, 2025 , 43(12) : 29 -37 . DOI: 10.3981/j.issn.1000-7857.2025.04.00082

1
Hughes J, Rees S, Kalindjian S, et al. Principles of early drug discovery[J]. British Journal of Pharmacology, 2011, 162(6): 1239- 1249.

DOI

2
Rake B, Gulbrandsen M, McKelvey M, et al. Rethinking medical innovation: Organizing R&D, responding to crisis, delivering health services[J]. Innovation, 2025, 27(1): 1- 20.

DOI

3
Lu M K, Yin J Y, Zhu Q, et al. Artificial intelligence in pharmaceutical sciences[J]. Engineering, 2023, 27: 37- 69.

DOI

4
Rehman A U, Li M Y, Wu B J, et al. Role of artificial intelligence in revolutionizing drug discovery[J]. Fundamental Research, 2025, 5(3): 1273- 1287.

DOI

5
Fu C, Chen Q C. The future of pharmaceuticals: Artificial intelligence in drug discovery and development[J]. Journal of Pharmaceutical Analysis, 2025, 1(7): 101248.

6
Roy R, Al-Hashimi H M. AlphaFold3 takes a step toward decoding molecular behavior and biological computation[J]. Nature Structural & Molecular Biology, 2024, 31(7): 997- 1000.

7
Abramson J, Adler J, Dunger J, et al. Accurate structure prediction of biomolecular interactions with AlphaFold3[J]. Nature, 2024, 630(8016): 493- 500.

DOI

8
He X H, Li J R, Shen S Y, et al. AlphaFold3 versus experimental structures: Assessment of the accuracy in ligand-bound G protein-coupled receptors[J]. Acta Pharmacologica Sinica, 2025, 46(4): 1111- 1122.

DOI

9
Baek M, DiMaio F, Anishchenko I, et al. Accurate prediction of protein structures and interactions using a three-track neural network[J]. Science, 2021, 373(6557): 871- 876.

DOI

10
Desai D, Kantliwala S V, Vybhavi J, et al. Review of AlphaFold3: Transformative advances in drug design and therapeutics[J]. Cureus, 2024, 16(7): e63646.

11
Chen E, Pan E, Zhang S G. Structure bioinformatics of six human integral transmembrane enzymes and their AlphaFold3 predicted water-soluble QTY analogs: Insights into FACE1 and STEA4 binding mechanisms[J]. Pharmaceutical Research, 2025, 42(2): 291- 305.

DOI

12
Wee J, Wei G W. Evaluation of AlphaFold3's protein-protein complexes for predicting binding free energy changes upon mutation[J]. Journal of Chemical Information and Modeling, 2024, 64(16): 6676- 6683.

DOI

13
Cobb R E, Chao R, Zhao H M. Directed evolution: Past, present, and future[J]. AIChE Journal, 2013, 59(5): 1432- 1440.

DOI

14
Strokach A, Kim P M. Deep generative modeling for protein design[J]. Current Opinion in Structural Biology, 2022, 72: 226- 236.

DOI

15
Dauparas J, Anishchenko I, Bennett N, et al. Robust deep learning-based protein sequence design using ProteinMPNN[J]. Science, 2022, 378(6615): 49- 56.

DOI

16
Yu T H, Cui H Y, Li J C, et al. Enzyme function prediction using contrastive learning[J]. Science, 2023, 379(6639): 1358- 1363.

DOI

17
Watson J L, Juergens D, Bennett N R, et al. De novo design of protein structure and function with RFdiffusion[J]. Nature, 2023, 620(7976): 1089- 1100.

DOI

18
Su X, Li J, Xu X, et al. Strategies to enhance the therapeutic efficacy of anti-PD-1 antibody, anti-PD-L1 antibody and anti-CTLA-4 antibody in cancer therapy[J]. Journal of Translational Medicine, 2024, 22(1): 751.

DOI

19
Cheng J, Liang T J, Xie X Q, et al. A new era of antibody discovery: An in-depth review of AI-driven approaches[J]. Drug Discovery Today, 2024, 29(6): 103984.

DOI

20
Zhao W B, Luo X W, Tong F, et al. Improving antibody optimization ability of generative adversarial network through large language model[J]. Computational and Structural Biotechnology Journal, 2023, 21: 5839- 5850.

DOI

21
Cai H Y, Zhang Z B, Wang M K, et al. Pretrainable geometric graph neural network for antibody affinity maturation[J]. Nature Communications, 2024, 15(1): 7785.

DOI

22
Hie B L, Shanker V R, Xu D, et al. Efficient evolution of human antibodies from general protein language models[J]. Nature Biotechnology, 2024, 42(2): 275- 283.

DOI

23
Mason D M, Friedensohn S, Weber C R, et al. Optimization of therapeutic antibodies by predicting antigen specificity from antibody sequence via deep learning[J]. Nature Biomedical Engineering, 2021, 5(6): 600- 612.

DOI

24
Saka K, Kakuzaki T, Metsugi S, et al. Antibody design using LSTM based deep generative model from phage display library for affinity maturation[J]. Scientific Reports, 2021, 11(1): 5852.

DOI

25
Bennett N R, Watson J L, Ragotte R J, et al. Atomically accurate de novo design of antibodies with RFdiffusion[J]. bioRxiv, 2025, 2024.

26
Liang T J, Sun Z Y, Hines M G, et al. AI-based IsAb2.0 for antibody design[J]. Briefings in Bioinformatics, 2024, 25(5): bbae445.

DOI

27
Zheng J Y, Wang Y, Liang Q Y, et al. The application of machine learning on antibody discovery and optimization[J]. Molecules, 2024, 29(24): 5923.

DOI

28
Jiang X F, Luo S Z, Liao K B, et al. Artificial intelligence and automation to power the future of chemistry[J]. Cell Reports Physical Science, 2024, 5(7): 102049.

DOI

29
You Y J, Lai X, Pan Y, et al. Artificial intelligence in cancer target identification and drug discovery[J]. Signal Transduction and Targeted Therapy, 2022, 7(1): 156.

DOI

30
Zhang X J, Zhang O, Shen C, et al. Efficient and accurate large library ligand docking with KarmaDock[J]. Nature Computational Science, 2023, 3(9): 789- 804.

DOI

31
Lin H T, Huang Y F, Zhang O, et al. DiffBP: Generative diffusion of 3D molecules for target protein binding[J]. Chemical Science, 2025, 16(3): 1417- 1431.

DOI

32
Xiong G L, Wu Z X, Yi J C, et al. ADMETlab 2.0: An integrated online platform for accurate and comprehensive predictions of ADMET properties[J]. Nucleic Acids Research, 2021, 49(W1): 5- 14.

DOI

33
Baker F N, Chen Z Q, Adu-Ampratwum D, et al. RLSynC: Offline-online reinforcement learning for synthon completion[J]. Journal of Chemical Information and Modeling, 2024, 64(17): 6723- 6735.

DOI

34
Zhavoronkov A, Ivanenkov Y A, Aliper A, et al. Deep learning enables rapid identification of potent DDR1 kinase inhibitors[J]. Nature Biotechnology, 2019, 37(9): 1038- 1040.

35
Vamathevan J, Clark D, Czodrowski P, et al. Applications of machine learning in drug discovery and development[J]. Nature Reviews Drug Discovery, 2019, 18(6): 463- 477.

36
Jiménez-Luna J, Grisoni F, Schneider G. Drug discovery with explainable artificial intelligence[J]. Nature Machine Intelligence, 2020, 2(10): 573- 584.

37
Li X Y, Xu Y Q, Yao H Q, et al. Chemical space exploration based on recurrent neural networks: Applications in discovering kinase inhibitors[J]. Journal of Cheminformatics, 2020, 12(1): 42.

38
Zhang P, Zhang D F, Zhou W A, et al. Network pharmacology: Towards the artificial intelligence-based precision traditional Chinese medicine[J]. Briefings in Bioinformatics, 2023, 25(1): bbad518.

39
Himmelstein D S, Lizee A, Hessler C, et al. Systematic integration of biomedical knowledge prioritizes drugs for repurposing[J]. eLife, 2017, 6: e26726.

40
Steiner S, Wolf J, Glatzel S, et al. Organic synthesis in a modular robotic system driven by a chemical programming language[J]. Science, 2019, 363(6423): eaav2211.

41
Taherdoost H, Ghofrani A. AI's role in revolutionizing personalized medicine by reshaping pharmacogenomics and drug therapy[J]. Intelligent Pharmacy, 2024, 2(5): 643- 650.

Outlines

/