Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Oct 7:e08896.
doi: 10.1002/advs.202508896. Online ahead of print.

B-EPIC: A Transformer-Based Language Model for Decoding B Cell Immunodominance Patterns

Jun-Ze Liang  1 , Youtao Wang  1 , Cong Sun  1 , Tao Liu  1 , Zengfeng Wu  1 , Lipeng Chen  1 , Lina Chen  1 , Penglin Li  1 , Zhengkang Li  2 , Cangui Zhang  3 , Bingyun Lu  3 , Ye Chen  3 , Bing Gu  2 , Qian Zhong  1 , Xin Wei Wang  4   5 , Mu-Sheng Zeng  1 , Jinping Liu  1
Affiliations

B-EPIC: A Transformer-Based Language Model for Decoding B Cell Immunodominance Patterns

Jun-Ze Liang et al. Adv Sci (Weinh). .

Abstract

Vaccine development for pathogens has faced significant challenges, contributing to a public health burden. B-cell epitope (BCE) prediction is a crucial process in vaccine development, but is hindered by limited efficiency and accuracy. To address this, B-Epic, the first pipeline applying Transformer to predict BCEs is independently developed. B-Epic's robustness is validated through multiple testing datasets, including distinguishing clinically-approved vaccine targets, identifying BCEs (the Immune Epitope Database testing dataset; n = 23,888) and immunoreactive peptides (Trypanosoma cruzi peptidome; n = 239,575) with high AUCs of 0.882 and 0.945, respectively, outperforming widely used tools. Based on its superior performance, B-Epic is applied to the prevention of carcinogenic pathogens. In the application to Helicobacter pylori, peptides screened by B-Epic can activate B cells in experiments, suggesting their potential as vaccine targets. In another application to Epstein-Barr virus, B-Epic identifies pan-immunoreactive peptides in a clinical cohort (n = 899). These peptides exhibit higher reactogenicity in nasopharyngeal carcinoma patients than in healthy controls (n = 140), indicating their viability as immunodiagnostic targets. Overall, B-Epic utilizes self-attention, high-dimensional feature projection, and convolutional neural networks to autonomously extract complicated BCE features, enabling accurate BCE prediction and thereby facilitating efforts to prevent infectious diseases and cancers.

Keywords: B cell epitope prediction; Immunodiagnostics design; pathogens prevention; transformer; vaccines development.

PubMed Disclaimer

References

    1. P. Bonanni, C. Sacco, R. Donato, R. Capei, Clin Microbiol. Infect. 2014, 20, 32.
    1. B. Greenwood, Philos. Trans. R. Soc. B: Biol. Sci. 2014, 369, 20130433.
    1. I. Mansoor, H. A. Eassa, K. H. A. Mohammed, M. A. Abd El‐Fattah, M. H. Abdo, E. Rashad, H. A. Eassa, A. Saleh, O. M. Amin, M. I. Nounou, O. Ghoneim, AAPS PharmSciTech 2022, 23, 103.
    1. Y. Li, X. Wang, D. M. Blau, M. T. Caballero, D. R. Feikin, C. J. Gill, S. A. Madhi, S. B. Omer, E. A. F. Simões, H. Campbell, A. B. Pariente, D. Bardach, Q. Bassat, J.‐S. Casalegno, G. Chakhunashvili, N. Crawford, D. Danilenko, L. A. H. Do, M. Echavarria, A. Gentile, A. Gordon, T. Heikkinen, Q. S. Huang, S. Jullien, A. Krishnan, E. L. Lopez, J. Markic, A. Mira‐Iglesias, H. C. Moore, J. Moyes, et al., Lancet 2022, 399, 2047.
    1. C. Sun, Y.‐F. Kang, X.‐Y. Fang, Y.‐N. Liu, G.‐L. Bu, A.‐J. Wang, Y. Li, Q.‐Y. Zhu, H. Zhang, C. Xie, X.‐W. Kong, Y.‐J. Peng, W.‐J. Lin, L. Zhou, X.‐C. Chen, Z.‐Z. Lu, H.‐Q. Xu, D.‐C. Hong, X. Zhang, L. Zhong, G.‐K. Feng, Y.‐X. Zeng, M. Xu, Q. Zhong, Z. Liu, M.‐S. Zeng, Cell Host Microbe 2023, 31, 1882.