Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Harnessing deep statistical potential for biophysical scoring of protein-peptide interactions

Acta Pharmacologica Sinica (2025)Cite this article

Abstract

Protein-peptide interactions (PpIs) play a critical role in major cellular processes. Recently, a number of machine learning (ML)-based methods have been developed to predict PpIs, but most of them rely heavily on sequence data, limiting their ability to capture the generalized molecular interactions in three-dimensional (3D) space, which is crucial for understanding protein-peptide binding mechanisms and advancing peptide therapeutics. Protein-peptide docking approaches provide a feasible way to generate the 3D models of PpIs, but they often suffer from low-precision scoring functions (SFs). To address this, we developed DeepPpIScore, a novel SF for PpIs that employs unsupervised geometric deep learning coupled with a physics-inspired statistical potential. Trained solely on curated experimental structures without binding affinity data or classification labels, DeepPpIScore exhibits broad generalization across multiple tasks. Our comprehensive evaluations in bound and unbound peptide bioactive conformation prediction, binding affinity prediction, and binding pair identification reveal that DeepPpIScore outperforms or matches state-of-the-art baselines, including popular protein-protein SFs, ML-based methods, and AlphaFold-Multimer 2.3 (AF-M 2.3). Notably, DeepPpIScore achieves superior results in peptide binding mode prediction compared to AF-M 2.3. More importantly, DeepPpIScore offers interpretability in terms of hotspot preferences at protein interfaces, physics-informed noncovalent interactions, and protein-peptide binding energies.

Access through your institution
Buy or subscribe

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: The workflow of DeepPpIScore.
Fig. 2: Performance evaluation of different scoring functions across six docking programs on the PepSet benchmark.
Fig. 3: Evaluation of DeepPpIScore and docking programs across varying peptide lengths and sequence identities on PepSet, with case studies.
Fig. 4: Evaluation of different scoring functions on the BoundPep benchmark.
Fig. 5: Performance comparison of ADCP, VoroMQA, DeepPpiScore, and AF-M 2.3 for scoring of protein-peptide pose.
Fig. 6: Evaluation of scoring functions for binding affinity prediction on two distinct protein-peptide test sets.
Fig. 7: DeepPpIScore identifies interface hot spots and correlates with experimental binding indicators at residue resolution.

Similar content being viewed by others

Data availability

The source data and main scripts that implement the computational protocol are available at https://github.com/zjujdj/DeepPpIScore.

Code availability

The source data and main scripts that implement the computational protocol are available at https://github.com/zjujdj/DeepPpIScore.

References

  1. Cunningham AD, Qvit N, Mochly-Rosen D. Peptides and peptidomimetics as regulators of protein–protein interactions. Curr Opin Struct Biol. 2017;44:59–66.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Venkatesan K, Rual J-F, Vazquez A, Stelzl U, Lemmens I, Hirozane-Kishikawa T, et al. An empirical framework for binary interactome mapping. Nat methods. 2009;6:83–90.

    Article  CAS  PubMed  Google Scholar 

  3. Stumpf MP, Thorne T, De Silva E, Stewart R, An HJ, Lappe M, et al. Estimating the size of the human interactome. Proc Natl Acad Sci. 2008;105:6959–64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Yin S, Mi X, Shukla D. Leveraging machine learning models for peptide–protein interaction prediction. RSC Chem Biol. 2024;5:401–17.

  5. Bruzzoni-Giovanelli H, Alezra V, Wolff N, Dong C-Z, Tuffery P, Rebollo A. Interfering peptides targeting protein–protein interactions: the next generation of drugs?. Drug Discov Today. 2018;23:272–85.

    Article  CAS  PubMed  Google Scholar 

  6. Muttenthaler M, King GF, Adams DJ, Alewood PF. Trends in peptide drug discovery. Nat Rev Drug Discov. 2021;20:309–25.

    Article  CAS  PubMed  Google Scholar 

  7. Martins PM, Santos LH, Mariano D, Queiroz FC, Bastos LL, Gomes IdS, et al. Propedia: a database for protein–peptide identification based on a hybrid clustering algorithm. BMC Bioinf. 2021;22:1.

  8. Lee H, Heo L, Lee MS, Seok C. GalaxyPepDock: a protein–peptide docking tool based on interaction similarity and energy optimization. Nucleic Acids Res. 2015;43:W431–W5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Xu X, Yan C, Zou X. MDockPeP: An ab-initio protein-peptide docking server. J Comput Chem. 2018;39:2409–13.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Alam N, Goldstein O, Xia B, Porter KA, Kozakov D, Schueler-Furman O. High-resolution global peptide-protein docking using fragments-based PIPER-FlexPepDock. PLoS Comput Biol. 2017;13:e1005905.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Zhou P, Jin B, Li H, Huang S-Y. HPEPDOCK: a web server for blind peptide–protein docking based on a hierarchical algorithm. Nucleic Acids Res. 2018;46:W443–W50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. 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;31:455–61.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Trellet M, Melquiond AS, Bonvin AM. A unified conformational selection and induced fit approach to protein-peptide docking. PLoS One. 2013;8:e58769.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Zhang Y, Sanner MF. AutoDock CrankPep: combining folding and docking to predict protein–peptide complexes. Bioinformatics. 2019;35:5121–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Mondal A, Chang L, Perez A. Modelling peptide–protein complexes: docking, simulations and machine learning. QRB Discov. 2022;3:e17.

  16. Yan C, Xu X, Zou X. Fully blind docking at the atomic level for protein-peptide complex structure prediction. Structure. 2016;24:1842–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Shanker S, Sanner MF. Predicting protein-peptide interactions: benchmarking deep learning techniques and a comparison with focused docking. J Chem Inf Model. 2023;63:3158–70.

    Article  CAS  PubMed  Google Scholar 

  18. Ciemny M, Kurcinski M, Kamel K, Kolinski A, Alam N, Schueler-Furman O, et al. Protein–peptide docking: opportunities and challenges. Drug Discov Today. 2018;23:1530–7.

    Article  CAS  PubMed  Google Scholar 

  19. Jiang D, Hsieh C-Y, Wu Z, Kang Y, Wang J, Wang E, et al. InteractionGraphNet: a novel and efficient deep graph representation learning framework for accurate protein–ligand interaction predictions. J Med Chem. 2021;64:18209–32.

  20. Jiang D, Ye Z, Hsieh C-Y, Yang Z, Zhang X, Kang Y, et al. MetalProGNet: a structure-based deep graph model for metalloprotein–ligand interaction predictions. Chem Sci. 2023;14:2054–69.

  21. Senior AW, Evans R, Jumper J, Kirkpatrick J, Sifre L, Green T, et al. Improved protein structure prediction using potentials from deep learning. Nature. 2020;577:706–10.

    Article  CAS  PubMed  Google Scholar 

  22. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596:583–9.

  23. Tsaban T, Varga JK, Avraham O, Ben-Aharon Z, Khramushin A, Schueler-Furman O. Harnessing protein folding neural networks for peptide-protein docking. Nat Commun. 2022;13:176.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Evans R, O’Neill M, Pritzel A, Antropova N, Senior A, Green T, et al. Protein complex prediction with AlphaFold-Multimer. bioRxiv 2022: 2021.10.04.463034.

  25. Johansson-Akhe I, Wallner B. Improving peptide-protein docking with AlphaFold-Multimer using forced sampling. Front Bioinforma. 2022;2:959160.

    Article  Google Scholar 

  26. Johansson-Akhe I, Mirabello C, Wallner B. InterPep2: global peptide-protein docking using interaction surface templates. Bioinformatics. 2020;36:2458–65.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Kurcinski M, Jamroz M, Blaszczyk M, Kolinski A, Kmiecik S. CABS-dock web server for the flexible docking of peptides to proteins without prior knowledge of the binding site. Nucleic Acids Res. 2015;43:W419–W24.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Pierce BG, Wiehe K, Hwang H, Kim B-H, Vreven T, Weng Z. ZDOCK server: interactive docking prediction of protein–protein complexes and symmetric multimers. Bioinformatics. 2014;30:1771–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Abdin O, Nim S, Wen H, Kim PM. PepNN: a deep attention model for the identification of peptide binding sites. Commun Biol. 2022;5:503.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Wang R, Jin J, Zou Q, Nakai K, Wei L. Predicting protein–peptide binding residues via interpretable deep learning. Bioinformatics. 2022;38:3351–60.

    Article  CAS  PubMed  Google Scholar 

  31. Lei Y, Li S, Liu Z, Wan F, Tian T, Li S, et al. A deep-learning framework for multi-level peptide–protein interaction prediction. Nat Commun. 2021;12:5465.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Taherzadeh G, Zhou Y, Liew AW-C, Yang Y. Structure-based prediction of protein–peptide binding regions using Random Forest. Bioinformatics. 2018;34:477–84.

    Article  CAS  PubMed  Google Scholar 

  33. Johansson-Åkhe I, Mirabello C, Wallner B. Predicting protein-peptide interaction sites using distant protein complexes as structural templates. Sci Rep -Uk. 2019;9:4267.

    Article  Google Scholar 

  34. Johansson-Åkhe I, Mirabello C, Wallner B. InterPepRank: assessment of docked peptide conformations by a deep graph network. Front Bioinf. 2021;1:763102.

  35. Johansson-Akhe I, Wallner B. InterPepScore: a deep learning score for improving the FlexPepDock refinement protocol. Bioinformatics. 2022;38:3209–15.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Shen C, Zhang X, Deng Y, Gao J, Wang D, Xu L, et al. Boosting protein-ligand binding pose prediction and virtual screening based on residue-atom distance likelihood potential and graph transformer. J Med Chem. 2022;65:10691–706.

  37. Zhang X, Zhang O, Shen C, Qu W, Chen S, Cao H, et al. Efficient and accurate large library ligand docking with KarmaDock. Nat Comput. Sci. 2023;3:789–804.

    Article  CAS  PubMed  Google Scholar 

  38. Méndez-Lucio O, Ahmad M, del Rio-Chanona EA, Wegner JK. A geometric deep learning approach to predict binding conformations of bioactive molecules. Nat Mach Intell. 2021;3:1033–9.

    Article  Google Scholar 

  39. Weng G, Gao J, Wang Z, Wang E, Hu X, Yao X, et al. Comprehensive Evaluation of Fourteen Docking Programs on Protein-Peptide Complexes. J Chem Theory Comput. 2020;16:3959–69.

    Article  CAS  PubMed  Google Scholar 

  40. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, et al. The protein data bank. Nucleic Acids Res. 2000;28:235–42.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Wang R, Fang X, Lu Y, Yang C-Y, Wang S. The PDBbind database: methodologies and updates. J Med Chem. 2005;48:4111–9.

    Article  CAS  PubMed  Google Scholar 

  42. Ge J, Jiang D, Sun H, Kang Y, Pan P, Deng Y, et al. Deep-learning-based prediction framework for protein-peptide interactions with structure generation pipeline. Cell Rep Phys Sci. 2024;5:101980.

  43. Wen Z, He J, Tao H, Huang S-Y. Pepbdb: a comprehensive structural database of biological peptide–protein interactions. Bioinformatics. 2019;35:175–7.

    Article  CAS  PubMed  Google Scholar 

  44. Madhavi Sastry G, Adzhigirey M, Day T, Annabhimoju R, Sherman W. Protein and ligand preparation: parameters, protocols, and influence on virtual screening enrichments. J Comput-aided Mol Des. 2013;27:221–34.

    Article  CAS  PubMed  Google Scholar 

  45. Zhang C, Liu S, Zhu Q, Zhou Y. A knowledge-based energy function for protein− ligand, protein− protein, and protein− DNA complexes. J Med Chem. 2005;48:2325–35.

    Article  CAS  PubMed  Google Scholar 

  46. Yang H, Xiong Z, Zonta F. Construction of a deep neural network energy function for protein physics. J Chem Theory Comput. 2022;18:5649–58.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Fan H, Schneidman-Duhovny D, Irwin JJ, Dong G, Shoichet BK, Sali A. Statistical potential for modeling and ranking of protein-ligand interactions. J Chem Inf Model. 2011;51:3078–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Zhang L, Liu H. Exploring binding positions and backbone conformations of peptide ligands of proteins with a backbone-centred statistical energy function. J Comput-aided Mol Des. 2023;37:463–78.

    Article  CAS  PubMed  Google Scholar 

  49. Zhou H, Zhou Y. Distance-scaled, finite ideal-gas reference state improves structure-derived potentials of mean force for structure selection and stability prediction. Protein Sci. 2002;11:2714–26.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. Yang J, Anishchenko I, Park H, Peng Z, Ovchinnikov S, Baker D. Improved protein structure prediction using predicted interresidue orientations. Proc Natl Acad Sci. 2020;117:1496–503.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Lin Z, Akin H, Rao R, Hie B, Zhu Z, Lu W, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science. 2023;379:1123–30.

    Article  CAS  PubMed  Google Scholar 

  52. Jing B, Eismann S, Suriana P, Townshend RJL, Dror R. International Conference on Learning Representations.

  53. Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. 31st Annual Conference on Neural Information Processing Systems (NIPS).

  54. Feng T, Chen F, Kang Y, Sun H, Liu H, Li D, et al. HawkRank: a new scoring function for protein–protein docking based on weighted energy terms. J Cheminf. 2017;9:1–15.

    Article  CAS  Google Scholar 

  55. Pierce B, Weng Z. ZRANK: reranking protein docking predictions with an optimized energy function. Proteins: Struct Funct Bioinforma. 2007;67:1078–86.

    Article  CAS  Google Scholar 

  56. Andrusier N, Nussinov R, Wolfson HJ. FireDock: fast interaction refinement in molecular docking. Proteins: Struct Funct Bioinforma. 2007;69:139–59.

    Article  CAS  Google Scholar 

  57. Pallara C, Jiménez-García B, Romero M, Moal IH, Fernández-Recio J. pyDock scoring for the new modeling challenges in docking: Protein–peptide, homo-multimers, and domain–domain interactions. Proteins: Struct Funct Bioinforma. 2016;85:487–96.

    Article  Google Scholar 

  58. Cheng TMK, Blundell TL, Fernandez-Recio J. pyDock: Electrostatics and desolvation for effective scoring of rigid-body protein–protein docking. Proteins: Struct, Funct, Bioinforma. 2007;68:503–15.

    Article  CAS  Google Scholar 

  59. Yang Y, Zhou Y. Specific interactions for ab initio folding of protein terminal regions with secondary structures. Proteins: Struct, Funct, Bioinforma. 2008;72:793–803.

    Article  CAS  Google Scholar 

  60. Olechnovic K, Venclovas C. VoroMQA: Assessment of protein structure quality using interatomic contact areas. Proteins: Struct Funct Bioinforma. 2017;85:1131–45.

    Article  CAS  Google Scholar 

  61. Olechnovič K, Venclovas Č. VoroMQA web server for assessing three-dimensional structures of proteins and protein complexes. Nucleic Acids Res. 2019;47:W437–W42.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Wang X, Flannery ST, Kihara D. Protein docking model evaluation by graph neural networks. Front Mol Biosci. 2021;8:647915.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Sun P, Qu J, Lyu X, Ling H, Tang Z. Graph attribute aggregation network with progressive margin folding. arXiv preprint arXiv:.05347 2019.

  64. Jurtz V, Paul S, Andreatta M, Marcatili P, Peters B, Nielsen M. NetMHCpan-4.0: improved peptide–MHC class I interaction predictions integrating eluted ligand and peptide binding affinity data. J Immunol. 2017;199:3360–8.

    Article  CAS  PubMed  Google Scholar 

  65. O’Donnell TJ, Rubinsteyn A, Laserson U. MHCflurry 2.0: improved pan-allele prediction of MHC class I-presented peptides by incorporating antigen processing. Cell Syst. 2020;11:42–8. e7.

    Article  PubMed  Google Scholar 

  66. Bogan AA, Thorn KS. Anatomy of hot spots in protein interfaces. J Mol Biol. 1998;280:1–9.

    Article  CAS  PubMed  Google Scholar 

  67. Adasme MF, Linnemann KL, Bolz SN, Kaiser F, Salentin S, Haupt VJ, et al. PLIP 2021: expanding the scope of the protein-ligand interaction profiler to DNA and RNA. Nucleic Acids Res. 2021;49:W530–W534.

  68. Weng G, Wang E, Chen F, Sun H, Wang Z, Hou T. Assessing the performance of MM/PBSA and MM/GBSA methods. 9. Prediction reliability of binding affinities and binding poses for protein–peptide complexes. Phys Chem Chem Phys. 2019;21:10135–45.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

This work was financially supported by National Natural Science Foundation of China (22307112), Young Scientists Fund of Natural Science Foundation of Hunan Province of China (2025JJ60651), the National Key R&D Program of China (2024YFA1307500), Postdoctoral Science Foundation of China (2022M722777), and Postdoctoral Fellowship Program of CPSF (GZB20230648, GZB20230657).

Author information

Author notes

  1. These authors contributed equally: De-jun Jiang, Hui-feng Zhao, Hong-yan Du.

Authors and Affiliations

  1. College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, 310058, China

    De-jun Jiang, Hui-feng Zhao, Hong-yan Du, Yu Kang, Pei-chen Pan, Zhen-xing Wu, O-din Zhang, Xiao-rui Wang, Ji-ke Wang, Yuan-sheng Huang, Yi-hao Zhao, Chang-Yu Hsieh & Ting-jun Hou

  2. Xiangya School of Pharmaceutical Sciences, Central South University, Changsha, 410004, China

    De-jun Jiang & Dong-sheng Cao

  3. Hangzhou Carbonsilicon AI Technology Co., Ltd, Hangzhou, 310018, China

    De-jun Jiang

  4. College of Control Science and Engineering, Zhejiang University, Hangzhou, 310058, China

    Yun-dian Zeng

  5. Department of Medicinal Chemistry, China Pharmaceutical University, Nanjing, 210009, China

    Hui-yong Sun

Authors
  1. De-jun Jiang
    View author publications

    Search author on:PubMed Google Scholar

  2. Hui-feng Zhao
    View author publications

    Search author on:PubMed Google Scholar

  3. Hong-yan Du
    View author publications

    Search author on:PubMed Google Scholar

  4. Yu Kang
    View author publications

    Search author on:PubMed Google Scholar

  5. Pei-chen Pan
    View author publications

    Search author on:PubMed Google Scholar

  6. Zhen-xing Wu
    View author publications

    Search author on:PubMed Google Scholar

  7. Yun-dian Zeng
    View author publications

    Search author on:PubMed Google Scholar

  8. O-din Zhang
    View author publications

    Search author on:PubMed Google Scholar

  9. Xiao-rui Wang
    View author publications

    Search author on:PubMed Google Scholar

  10. Ji-ke Wang
    View author publications

    Search author on:PubMed Google Scholar

  11. Yuan-sheng Huang
    View author publications

    Search author on:PubMed Google Scholar

  12. Yi-hao Zhao
    View author publications

    Search author on:PubMed Google Scholar

  13. Chang-Yu Hsieh
    View author publications

    Search author on:PubMed Google Scholar

  14. Dong-sheng Cao
    View author publications

    Search author on:PubMed Google Scholar

  15. Hui-yong Sun
    View author publications

    Search author on:PubMed Google Scholar

  16. Ting-jun Hou
    View author publications

    Search author on:PubMed Google Scholar

Contributions

DJJ, HYD, and HFZ contributed to the main code and wrote the manuscript. DJJ performed the experiment. YDZ, ODZ and ZXW provided partial codes of this work. HFZ and XRW helped perform the analysis with constructive discussions. YDZ, JKW, YHZ, and YSH contributed to the visualization and technique support. YK, PCP, HYS, DSC, TJH, and CYH provided essential financial support and conception, and were responsible for the overall quality.

Corresponding authors

Correspondence to Chang-Yu Hsieh, Dong-sheng Cao, Hui-yong Sun or Ting-jun Hou.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jiang, Dj., Zhao, Hf., Du, Hy. et al. Harnessing deep statistical potential for biophysical scoring of protein-peptide interactions. Acta Pharmacol Sin (2025). https://doi.org/10.1038/s41401-025-01659-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41401-025-01659-8

Key words

Search

Advanced search

Quick links