Applications of Deep Learning Algorithms in Protein Structure Prediction and Design

Authors

  • Yijie He College of Health Science and Environmental Engineering, Shenzhen Technology University, Shenzhen, China

DOI:

https://doi.org/10.54097/aeebgv76

Keywords:

Deep Learning; Protein Structure Prediction; Antibody Design.

Abstract

The three-dimensional configuration of a protein is important for its biological function, which is crucial in biological study and drug development. The progress of deep learning has greatly improved the prediction of protein structures. Especially, the AlphaFold series has increased the prediction precision and extended its use in the simulation of biomolecular complexes and drug discovery. This paper describes the latest progress in protein structure prediction and antibody design based on deep learning. The emphasis is on the enhancement and efficiency of the template-based method and the AlphaFold series. Furthermore, the practical applications of these methods in antibody engineering and drug discovery are illustrated by considering specific examples such as AbEgDiffuser for simultaneous sequence and structure design, and AlphaFold3 for the choice of covalent ligands. Experimental data indicate that deep learning can precisely forecast the protein structures and effectively improve the design and screening of potential substances, thereby offering new possibilities for rational drug design. In the future, more research should concentrate on predicting the conformational changes of proteins, designing novel antibody frameworks from scratch, and obtaining multiple drug-like properties at the same time.

Downloads

Download data is not yet available.

References

[1] Bertoline L M F, Lima A N, Krieger J E, Teixeira S K. Before and after AlphaFold2: an overview of protein structure prediction[J]. Frontiers in Bioinformatics, 2023, 3: 1120370. DOI: https://doi.org/10.3389/fbinf.2023.1120370

[2] Almagro J C, Daniels-Wells T R, Perez-Tapia S M, Penichet M L. Progress and challenges in the design and clinical development of antibodies for cancer therapy[J]. Frontiers in Immunology, 2018, 8: 1751. DOI: https://doi.org/10.3389/fimmu.2017.01751

[3] 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: 871–876. https://doi.org/10.1126/science.abj8754. DOI: https://doi.org/10.1126/science.abj8754

[4] Watson J L, Juergens D, Bennett N R, et al. De novo design of protein structure and function with RFdiffusion[J]. Nature, 2023, 620: 1089–1100. https://doi.org/10.1038/s41586-023-06415-8. DOI: https://doi.org/10.1038/s41586-023-06415-8

[5] Waterhouse A, Bertoni M, Bienert S, et al. SWISS-MODEL: homology modelling of protein structures and complexes[J]. Nucleic Acids Research, 2018, 46(W1): W296–W303. DOI: https://doi.org/10.1093/nar/gky427

[6] McGuffin L J, Jones D T. Improvement of the GenTHREADER method for genomic fold recognition[J]. Bioinformatics, 2003, 19(7): 874–881. https://doi.org/10.1093/bioinformatics/btg097. DOI: https://doi.org/10.1093/bioinformatics/btg097

[7] Evans R, O’Neill M, Pritzel A, et al. Protein complex prediction with AlphaFold-Multimer[J]. bioRxiv, 2021: 2021.10.04.463034. DOI: https://doi.org/10.1101/2021.10.04.463034

[8] Yin R, Feng B Y, Varshney A, Pierce B G. Benchmarking AlphaFold for protein complex modeling reveals accuracy determinants[J]. Protein Science, 2022, 31(8): e4379. https://doi.org/10.1002/pro.4379. DOI: https://doi.org/10.1002/pro.4379

[9] Jumper J, Evans R, Pritzel A, et al. Highly accurate protein structure prediction with AlphaFold[J]. Nature, 2021, 596(7873): 583–589. DOI: https://doi.org/10.1038/s41586-021-03819-2

[10] Tunyasuvunakool K, Adler J, Wu Z, et al. Highly accurate protein structure prediction for the human proteome[J]. Nature, 2021, 596: 590–596. https://doi.org/10.1038/s41586-021-03828-1. DOI: https://doi.org/10.1038/s41586-021-03828-1

[11] Abramson J, Adler J, Dunger J, et al. Accurate structure prediction of biomolecular interactions with AlphaFold 3[J]. Nature, 2024, 630(8016): 493–500. DOI: https://doi.org/10.1038/s41586-024-07487-w

[12] Zhu Y, Shi X, Zhang J, Sun W, Wang L. AbEgDiffuser: antibody sequence–structure co-design with equivariant graph neural networks and diffusion models[J]. Journal of Chemical Theory and Computation, 2025, 21(21): 11307–11317. https://doi.org/10.1021/acs.jctc.5c00990. DOI: https://doi.org/10.1021/acs.jctc.5c00990

[13] Shamir Y, Gabizon R, Rogel A, Lin D Y, Andreotti A H, London N. Discovery of covalent ligands with AlphaFold3[J]. Journal of the American Chemical Society, 2026. https://doi.org/10.1021/jacs.5c22222 DOI: https://doi.org/10.1101/2025.03.19.642201

Downloads

Published

15-11-2026

Issue

Section

Articles

How to Cite

He, Y. (2026). Applications of Deep Learning Algorithms in Protein Structure Prediction and Design. International Journal of Biology and Life Sciences, 17(2), 8-14. https://doi.org/10.54097/aeebgv76