A protein–ligand binding affinity prediction model integrating multi-scale interaction features_Journal of Biomedical Engineering

Authors：

LIU Hui ^1,2,3 , YAN Lixin ¹ , WANG Sushan ¹ ,  PENG Jiansheng ^4,5

1. School of Computer Science and Technology, Dalian University of Technology, Dalian, Liaoning 116081, P. R. China;
2. School of Control Science and Engineering, Dalian University of Technology, Dalian, Liaoning 116081, P. R. China;
3. Key Laboratory of Integrated Circuits and Biomedical Electronic Systems of Liaoning Province, Dalian, Liaoning 116081, P. R. China;
4. School of Artificial Intelligence and Manufacturing, Hechi University, Yizhou, Guangxi 546300, P. R. China;
5. Key Laboratory of Artificial Intelligence and Information Processing of Guangxi Higher Education Institutions, Yizhou, Guangxi 546300, P. R. China;

Corresponding?author：

PENG Jiansheng, Email: sheng120410@163.com

Keywords：

Drug design; Binding affinity prediction; Feature fusion; Attention mechanism; Deep learning

DOI：

10.7507/1001-5515.202507045

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Protein-ligand binding affinity prediction is critical for drug development, where rapid and accurate assessment of drug-target binding remains a key challenge. Deep learning-based modeling provides a more efficient and scalable solution compared to traditional experimental approaches. This paper proposes a multi-scale interaction feature fusion model for protein–ligand binding affinity prediction, named the multi-scale binding affinity predictor (MSBind). The MSBind model employed multiple attention mechanisms to extract both global and local interaction features between the protein-ligand complex. By jointly modeling this multi-scale information, it ultimately enhanced the model’s predictive performance. Experimental results showed that on the PDBbind v2016 core dataset, the root mean square error of MSBind was only 1.23, which was significantly lower than the baselines, demonstrating superior predictive accuracy. Furthermore, dimensionality reduction visualization of the extracted interaction features provided additional validation of the interpretability and effectiveness of MSBind. In summary, by integrating multi-scale interaction features, MSBind provides a high-performance solution for protein-ligand binding affinity prediction.

Citation： LIU Hui, YAN Lixin, WANG Sushan, PENG Jiansheng. A protein–ligand binding affinity prediction model integrating multi-scale interaction features. Journal of Biomedical Engineering, 2026, 43(2): 337-343. doi: 10.7507/1001-5515.202507045 Copy

1.	曹雨康, 江健, 劉杰. 評分函數在蛋白質-配體結合方面的應用研究進展. 計算機應用與軟件, 2023, 40(7): 61-70.
2.	吳梅花, 陳小娥, 林琳, 等. 人工智能技術在藥物設計領域的應用與展望. 中國處方藥, 2025, 23(24): 120-123.
3.	Konc J, Le?nik S, Jane?i? D. Modeling enzyme-ligand binding in drug discovery. J Cheminform, 2015, 7: 48.
4.	Dhakal A, McKay C, Tanner J J, et al. Artificial intelligence in the prediction of protein-ligand interactions: recent advances and future directions. Brief Bioinform, 2022, 23(1): bbab476.
5.	Bag A, Ghorai P Kr. Development of quantum chemical method to calculate half maximal inhibitory concentration (IC50). Mol Inform, 2016, 35(5): 199-206.
6.	Torreri P, Ceccarini M, Macioce P, et al. Biomolecular interactions by surface plasmon resonance technology. Ann Ist Super Sanita, 2005, 41(4): 437-441.
7.	Leavitt S, Freire E. Direct measurement of protein binding energetics by isothermal titration calorimetry. Curr Opin Struc Biol, 2001, 11(5): 560-566.
8.	Mpofu K T, Ombinda-Lemboumba S, Mthunzi-Kufa P. Classical and quantum surface plasmon resonance biosensing. Int J Opt, 2023, 2023: 5538161.
9.	倪武衛, 曠文靜, 田介峰, 等. 計算化學在藥物化學中的應用. 天津藥學, 2025, 37(4): 503-512.
10.	Zhao J T, Cao Y, Zhang L. Exploring the computational methods for protein-ligand binding site prediction. Comput Struct Biotechnol J, 2020, 18: 417-426.
11.	Zhang W, Lin W R, Zhang D, et al. Recent advances in the machine learning-based drug-target interaction prediction. Curr Drug Metab, 2019, 20(3): 194-202.
12.	Ballester P J, Mitchell J B O. A machine learning approach to predicting protein–ligand binding affinity with applications to molecular docking. Bioinformatics, 2010, 26(9): 1169-1175.
13.	Kinnings S L, Liu N, Tonge P J, et al. A machine learning-based method to improve docking scoring functions and its application to drug repurposing. J Chem Inf Model, 2011, 51(2): 408-419.
14.	劉正美, 魏雪梅, 張俊鵬, 等. 藥物分子與靶標蛋白結合親和力預測研究進展. 計算機工程與應用, 2024, 60(23): 79-90.
15.	Wang H W. Prediction of protein-ligand binding affinity via deep learning models. Brief Bioinform, 2024, 25(2): bbae081.
16.	張民權, 龔銘城, 陳澤鍇, 等. 人工智能與分子模擬在藥物設計中的研究進展. 醫藥導報, 2024, 43(1): 78-84.
17.	Wang D D, Wu W H, Wang R. Structure-based, deep-learning models for protein-ligand binding affinity prediction. J Cheminform, 2024, 16(1): 2.
18.	Xia C Q, Feng S H, Xia Y, et al. Leveraging scaffold information to predict protein-ligand binding affinity with an empirical graph neural network. Brief Bioinform, 2023, 24(1): bbac603.
19.	Kyro G W, Brent R I, Batista V S. HAC-Net: a hybrid attention-based convolutional neural network for highly accurate protein–ligand binding affinity prediction. J Chem Inf Model, 2023, 63(7): 1947-1960.
20.	Zhang Y J, Li S Y, Meng K, et al. Machine learning for sequence and structure-based protein-ligand interaction prediction. J Chem Inf Model, 2024, 64(5): 1456-1472.
21.	Weininger D. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules. J Chem Inf Comput Sci, 1988, 28(1): 31-36.
22.	?ztürk H, ?zgür A, Ozkirimli E. DeepDTA: deep drug–target binding affinity prediction. Bioinformatics, 2018, 34(17): 821-829.
23.	Wang K L, Zhou R Y, Li Y H, et al. DeepDTAF: a deep learning method to predict protein-ligand binding affinity. Brief Bioinform, 2021, 22(5): bbab072.
24.	Abbasi K, Razzaghi P, Poso A, et al. DeepCDA: deep cross-domain compound–protein affinity prediction through LSTM and convolutional neural networks. Bioinformatics, 2020, 36(17): 4633-4642.
25.	Karimi M, Wu D, Wang Z, et al. DeepAffinity: interpretable deep learning of compound–protein affinity through unified recurrent and convolutional neural networks. Bioinformatics, 2019, 35(18): 3329-3338.
26.	Zhao L L, Zhu Y, Wen N F, et al. Drug-target binding affinity prediction in a continuous latent space using variational autoencoders. IEEE/ACM Trans Comput Biol Bioinform, 2024, 21(5): 1458-1467.
27.	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need// NIPS'17: Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: NIPS, 2017: 6000-6010.
28.	Le D P C, Wang D, Le V T. A comprehensive survey of recent transformers in image, video and diffusion models. CMC-Comput Mater Con, 2024, 80(1): 37-60.
29.	Abibullaev B, Keutayeva A, Zollanvari A. Deep learning in EEG-based BCIs: a comprehensive review of transformer models, advantages, challenges, and applications. IEEE Access, 2023, 11: 127271-127301.
30.	Ghimire A, Tayara H, Xuan Z Y, et al. CSatDTA: prediction of drug–target binding affinity using convolution model with self-attention. Int J Mol Sci, 2022, 23(15): 8453.
31.	Zhao Q, Duan G, Yang M, et al. AttentionDTA: drug–target binding affinity prediction by sequence-based deep learning with attention mechanism. IEEE/ACM Trans Comput Biol Bioinform, 2023, 20(2): 852-863.
32.	Jin Z, Wu T, Chen T, et al. CAPLA: improved prediction of protein–ligand binding affinity by a deep learning approach based on a cross-attention mechanism. Bioinformatics, 2023, 39(2): btad049.
33.	Wang M Y, Li W M, Jin Q. AffinityVAE: a multi-objective model for protein-ligand affinity prediction and drug design. Comput Biol Chem, 2023, 107: 107971.
34.	Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions. arXiv, 2016: 1511.07122.
35.	Wang R, Fang X, Lu Y, et al. The PDBbind database: collection of binding affinities for protein-ligand complexes with known three-dimensional structures. J Med Chem, 2004, 47(12): 2977-2980.
36.	O’Boyle N M, Banck M, James C A, et al. Open Babel: an open chemical toolbox. J Cheminform, 2011, 3(1): 33.
37.	Zhou P, Xie X, Lin Z, et al. Towards understanding convergence and generalization of AdamW. IEEE Trans Pattern Anal Mach Intell, 2024, 46(9): 6486-6493.
38.	Stepniewska-Dziubinska M M, Zielenkiewicz P, Siedlecki P. Development and evaluation of a deep learning model for protein–ligand binding affinity prediction. Bioinformatics, 2018, 34(21): 3666-3674.
39.	McInnes L, Healy J, Melville J. UMAP: uniform manifold approximation and projection for dimension reduction. arXiv, 2020: 1802.03426.
40.	Khan A K A, Malim N H A H. Comparative studies on resampling techniques in machine learning and deep learning models for drug-target interaction prediction. Molecules, 2023, 28(4): 1663.

1. 曹雨康, 江健, 劉杰. 評分函數在蛋白質-配體結合方面的應用研究進展. 計算機應用與軟件, 2023, 40(7): 61-70.
2. 吳梅花, 陳小娥, 林琳, 等. 人工智能技術在藥物設計領域的應用與展望. 中國處方藥, 2025, 23(24): 120-123.
3. Konc J, Le?nik S, Jane?i? D. Modeling enzyme-ligand binding in drug discovery. J Cheminform, 2015, 7: 48.
4. Dhakal A, McKay C, Tanner J J, et al. Artificial intelligence in the prediction of protein-ligand interactions: recent advances and future directions. Brief Bioinform, 2022, 23(1): bbab476.
5. Bag A, Ghorai P Kr. Development of quantum chemical method to calculate half maximal inhibitory concentration (IC50). Mol Inform, 2016, 35(5): 199-206.
6. Torreri P, Ceccarini M, Macioce P, et al. Biomolecular interactions by surface plasmon resonance technology. Ann Ist Super Sanita, 2005, 41(4): 437-441.
7. Leavitt S, Freire E. Direct measurement of protein binding energetics by isothermal titration calorimetry. Curr Opin Struc Biol, 2001, 11(5): 560-566.
8. Mpofu K T, Ombinda-Lemboumba S, Mthunzi-Kufa P. Classical and quantum surface plasmon resonance biosensing. Int J Opt, 2023, 2023: 5538161.
9. 倪武衛, 曠文靜, 田介峰, 等. 計算化學在藥物化學中的應用. 天津藥學, 2025, 37(4): 503-512.
10. Zhao J T, Cao Y, Zhang L. Exploring the computational methods for protein-ligand binding site prediction. Comput Struct Biotechnol J, 2020, 18: 417-426.
11. Zhang W, Lin W R, Zhang D, et al. Recent advances in the machine learning-based drug-target interaction prediction. Curr Drug Metab, 2019, 20(3): 194-202.
12. Ballester P J, Mitchell J B O. A machine learning approach to predicting protein–ligand binding affinity with applications to molecular docking. Bioinformatics, 2010, 26(9): 1169-1175.
13. Kinnings S L, Liu N, Tonge P J, et al. A machine learning-based method to improve docking scoring functions and its application to drug repurposing. J Chem Inf Model, 2011, 51(2): 408-419.
14. 劉正美, 魏雪梅, 張俊鵬, 等. 藥物分子與靶標蛋白結合親和力預測研究進展. 計算機工程與應用, 2024, 60(23): 79-90.
15. Wang H W. Prediction of protein-ligand binding affinity via deep learning models. Brief Bioinform, 2024, 25(2): bbae081.
16. 張民權, 龔銘城, 陳澤鍇, 等. 人工智能與分子模擬在藥物設計中的研究進展. 醫藥導報, 2024, 43(1): 78-84.
17. Wang D D, Wu W H, Wang R. Structure-based, deep-learning models for protein-ligand binding affinity prediction. J Cheminform, 2024, 16(1): 2.
18. Xia C Q, Feng S H, Xia Y, et al. Leveraging scaffold information to predict protein-ligand binding affinity with an empirical graph neural network. Brief Bioinform, 2023, 24(1): bbac603.
19. Kyro G W, Brent R I, Batista V S. HAC-Net: a hybrid attention-based convolutional neural network for highly accurate protein–ligand binding affinity prediction. J Chem Inf Model, 2023, 63(7): 1947-1960.
20. Zhang Y J, Li S Y, Meng K, et al. Machine learning for sequence and structure-based protein-ligand interaction prediction. J Chem Inf Model, 2024, 64(5): 1456-1472.
21. Weininger D. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules. J Chem Inf Comput Sci, 1988, 28(1): 31-36.
22. ?ztürk H, ?zgür A, Ozkirimli E. DeepDTA: deep drug–target binding affinity prediction. Bioinformatics, 2018, 34(17): 821-829.
23. Wang K L, Zhou R Y, Li Y H, et al. DeepDTAF: a deep learning method to predict protein-ligand binding affinity. Brief Bioinform, 2021, 22(5): bbab072.
24. Abbasi K, Razzaghi P, Poso A, et al. DeepCDA: deep cross-domain compound–protein affinity prediction through LSTM and convolutional neural networks. Bioinformatics, 2020, 36(17): 4633-4642.
25. Karimi M, Wu D, Wang Z, et al. DeepAffinity: interpretable deep learning of compound–protein affinity through unified recurrent and convolutional neural networks. Bioinformatics, 2019, 35(18): 3329-3338.
26. Zhao L L, Zhu Y, Wen N F, et al. Drug-target binding affinity prediction in a continuous latent space using variational autoencoders. IEEE/ACM Trans Comput Biol Bioinform, 2024, 21(5): 1458-1467.
27. Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need// NIPS'17: Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: NIPS, 2017: 6000-6010.
28. Le D P C, Wang D, Le V T. A comprehensive survey of recent transformers in image, video and diffusion models. CMC-Comput Mater Con, 2024, 80(1): 37-60.
29. Abibullaev B, Keutayeva A, Zollanvari A. Deep learning in EEG-based BCIs: a comprehensive review of transformer models, advantages, challenges, and applications. IEEE Access, 2023, 11: 127271-127301.
30. Ghimire A, Tayara H, Xuan Z Y, et al. CSatDTA: prediction of drug–target binding affinity using convolution model with self-attention. Int J Mol Sci, 2022, 23(15): 8453.
31. Zhao Q, Duan G, Yang M, et al. AttentionDTA: drug–target binding affinity prediction by sequence-based deep learning with attention mechanism. IEEE/ACM Trans Comput Biol Bioinform, 2023, 20(2): 852-863.
32. Jin Z, Wu T, Chen T, et al. CAPLA: improved prediction of protein–ligand binding affinity by a deep learning approach based on a cross-attention mechanism. Bioinformatics, 2023, 39(2): btad049.
33. Wang M Y, Li W M, Jin Q. AffinityVAE: a multi-objective model for protein-ligand affinity prediction and drug design. Comput Biol Chem, 2023, 107: 107971.
34. Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions. arXiv, 2016: 1511.07122.
35. Wang R, Fang X, Lu Y, et al. The PDBbind database: collection of binding affinities for protein-ligand complexes with known three-dimensional structures. J Med Chem, 2004, 47(12): 2977-2980.
36. O’Boyle N M, Banck M, James C A, et al. Open Babel: an open chemical toolbox. J Cheminform, 2011, 3(1): 33.
37. Zhou P, Xie X, Lin Z, et al. Towards understanding convergence and generalization of AdamW. IEEE Trans Pattern Anal Mach Intell, 2024, 46(9): 6486-6493.
38. Stepniewska-Dziubinska M M, Zielenkiewicz P, Siedlecki P. Development and evaluation of a deep learning model for protein–ligand binding affinity prediction. Bioinformatics, 2018, 34(21): 3666-3674.
39. McInnes L, Healy J, Melville J. UMAP: uniform manifold approximation and projection for dimension reduction. arXiv, 2020: 1802.03426.
40. Khan A K A, Malim N H A H. Comparative studies on resampling techniques in machine learning and deep learning models for drug-target interaction prediction. Molecules, 2023, 28(4): 1663.

Journal of Biomedical Engineering

A protein–ligand binding affinity prediction model integrating multi-scale interaction features

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content