| 27 | 0 | 38 |
| 下载次数 | 被引频次 | 阅读次数 |
B细胞线性表位的实验鉴定不仅成本高，且通量低；同时，已开发的相关预测模型也存在泛化能力不足的问题。为此，本研究提出了一种多深度学习算法融合的B细胞线性表位预测模型——DFPred,由CNN、单头注意力机制和xLSTM组成，并且在训练数据集和验证数据集上对DFPred的性能进行了广泛测试。实验结果表明，DFPred的AUC和AUC10%分别达到0.806和0.420,均优于对比方法。
Abstract：The experimental identification of linear B-cell epitopes is costly and low-throughput, while existing prediction models generally exhibit poor generalization ability. To address these challenges, this study proposes DFPred, a novel linear B-cell epitope prediction model that fuses multiple deep learning algorithms, specifically a convolutional neural network(CNN), a single-head attention mechanism, and extended long short-term memory(xLSTM). The performance of DFPred was extensively evaluated on both training and validation datasets. Experimental results demonstrate that DFPred achieved AUC and AUC10% values of 0.806 and 0.420, respectively, both of which outperformed the other comparative methods.
[1] 羊红光，成彬，王程.B细胞抗原表位预测[M].长春：吉林大学出版社，2025.
[2] El-MANZALAWY Y,HONAVAR V.Recent advances in B-cell epitope prediction methods[J].Immunome Research,2010,6(S2):S2.DOI:10.1186/1745-7580-6-S2-S2.
[3] TAREK A A,EWEIDA A E,SHEWEITA S A.B-cell epitope mapping for the design of vaccines and effective diagnostics[J].Trials in Vaccinology,2016,5:71-83.
[4] ASADI N,YOUSEFI E,FEIZOLLAHZADEH S,et al.Design of a multi-epitope antigen for toxoplasmosis diagnosis:an immunoinformatics approach[J].Acta Parasitologica,2025,70:192.DOI:10.1007/s11685-025-01132-w.
[5] MELADO L,KUMAR A,KALRA B,et al.How to enhance prediction of clinical outcomes in poor responders:integrating high-specific assays for anti-mullerian hormone with antral follicle count[J].Frontiers in Endocrinology,2025,16:1654365.DOI:10.3389/fendo.2025.1654365.
[6] MIRZAIE S,YUAN K D,NI H Y,et al.Design of a novel multiepitope vaccine against glioblastoma by in silico approaches[J].Scientific Reports,2025,15:24046.DOI:10.1038/s41598-025-03672-7.
[7] VIHINEN M,TORKKILA E,RIIKONEN P.Accuracy of protein flexibility predictions[J].Proteins,1994,19(2):141-149.
[8] PARKER J M,GUO D,HODGES R S.New hydrophilicity scale derived from high-performance liquid chromatography peptide retention data:correlation of predicted surface residues with antigenicity and X-ray-derived accessible sites[J].Biochemistry,1986,25(19):5425-5432.
[9] HOPP T P,WOODS K R.Prediction of protein antigenic determinants from amino acid sequences[J].Proceedings of the National Academy of Sciences of the United States of America,1981,78(6):3824-3828.
[10] LARSEN J E P,LUND O,NIELSEN M.Improved method for predicting linear B-cell epitopes[J].Immunome Research,2006,2:2.DOI:10.1186/1745-7580-2-2.
[11] ZHANG Wen,XIONG Yi,ZHAO Meng,et al.Prediction of conformational B-cell epitopes from 3D structures by random forests with a distance-based feature[J].BMC Bioinformatics,2011,12:341.DOI:10.1186/1471-2105-12-341.
[12] WANG H W,LIN Y C,PAI T W,et al.Prediction of B-cell linear epitopes with a combination of support vector machine classification and amino acid propensity identification[J].Journal of Biomedicine and Biotechnology,2011,2011:432830.DOI:10.1155/2011/432830.
[13] SAHA S,RAGHAVA G P S.Prediction of continuous B-cell epitopes in an antigen using recurrent neural network[J].Proteins-Structure Function and Bioinformatics,2006,65(1):40-48.
[14] COLLATZ M,MOCK F,BARTH E,et al.EpiDope:a deep neural network for linear B-cell epitope prediction[J].Bioinformatics,2021,37(12):1784.DOI:10.1093/bioinformatics/btab390.
[15] ALGHAMDI W,ATTIQUE M,ALZAHRANI E,et al.LBCEPred:a machine learning model to predict linear B-cell epitopes[J].Briefings in Bioinformatics,2022,23(3):bbac035.DOI:10.1093/bib/bbac035.
[16] CHUANG C C,LIU Y C,OU Y Y.DeepEpiIL13:deep learning for rapid and accurate prediction of IL-13-Inducing epitopes using pretrained language models and multiwindow convolutional neural networks[J].ACS Omega,2025,10(9):9675-9683.
[17] QI Yue,ZHENG Peijie,HUANG Guohua.DeepLBCEPred:a Bi-LSTM and multi-scale CNN-based deep learning method for predicting linear B-cell epitopes[J].Frontiers in Microbiology,2023,14:1117027.DOI:10.3389/fmicb.2023.1117027.
[18] DA SILVA B M,ASCHER D B,PIRES D E V.Epitope1D:accurate taxonomy-aware b-cell linear epitope prediction[J].Briefings in Bioinformatics,2023,24(3):bbad114.DOI:10.1093/bib/bbad114.
[19] HU Ruisi,GU Kui,EHSAN M,et al.Transformer-based deep learning enables improved B-cell epitope prediction in parasitic pathogens:a proof-of-concept study on Fasciola hepatica[J].PLoS Neglected Tropical Diseases,2025,19(4):e0012985.DOI:10.1371/journal.pntd.0012985.
[20] CHENG Bin,LIU Lingyun,QI Zhaohui,et al.Prediction of continuous B-cell epitopes using long short term memory networks[C]//In Proceedings of the 2018 6th International Conference on Bioinformatics and Computational Biology.New York,NY,USA:Association for Computing Machinery,2018:55-59.
[21] BECK M,P?PPEL K,SPANRING M,et al.Xlstm:extended long short-term memory[J].Advances in Neural Information Processing Systems,2024,37:107547-107603.
[22] LIN Zeming,AKIN H,RAO R,et al.Evolutionary-scale prediction of atomic-level protein structure with a language model[J].Science,2023,379(6637):1123-1130.
[23] SUN Jing,XU Tianlei,WANG Shuning,et al.Does difference exist between epitope and non-epitope residues?Analysis of the physicochemical and structural properties on conformational epitopes from B-cell protein antigens.[J].Immunome Research,2011,7(3):1-11.
基本信息:
DOI：10.16191/j.cnki.hbkx.2025.06.001
中图分类号:R392;TP18
引用信息:
[1]羊红光,苏柏馨.基于深度学习算法融合的B细胞线性表位预测模型[J].河北省科学院学报,2025,42(06):1-8.DOI:10.16191/j.cnki.hbkx.2025.06.001.
基金信息:
河北省自然科学基金项目(F2022302001)
