Bioinformatics. 2021 Jun 28; doi: 10.1093/bioinformatics/btab467. Epub 2021 Jun 28.
EpitopeVec: Linear Epitope Prediction Using Deep Protein Sequence Embeddings.
Bioinformatics (Oxford, England)
Akash Bahai, Ehsaneddin Asgari, Mohammad R K Mofrad, Andreas Kloetgen, Alice C McHardy
Affiliations
Affiliations
- Computational Biology of Infection Research, Helmholtz Center for Infection Research, 38124 Braunschweig, Germany.
- Braunschweig Integrated Center of Systems Biology (BRICS), Technische Universität Braunschweig, Rebenring 56, 38106 Braunschweig.
- Molecular Cell Biomechanics Laboratory, Departments of Bioengineering and Mechanical Engineering, University of California, Berkeley, CA, 94720, USA.
- Molecular Biophysics and Integrated Bioimaging, Lawrence Berkeley National Lab, Berkeley, CA 94720, USA.
PMID: 34180989
PMCID: PMC8652027 DOI: 10.1093/bioinformatics/btab467
Abstract
MOTIVATION: B-cell epitopes (BCEs) play a pivotal role in the development of peptide vaccines, immuno-diagnostic reagents, and antibody production, and thus in infectious disease prevention and diagnostics in general. Experimental methods used to determine BCEs are costly and time-consuming. Therefore, it is essential to develop computational methods for the rapid identification of BCEs. Although several computational methods have been developed for this task, generalizability is still a major concern, where cross-testing of the classifiers trained and tested on different datasets has revealed accuracies of 51-53.
RESULTS: We describe a new method called EpitopeVec, which uses a combination of residue properties, modified antigenicity scales, and protein language model-based representations (protein vectors) as features of peptides for linear BCE predictions. Extensive benchmarking of EpitopeVec and other state-of-the-art methods for linear BCE prediction on several large and small datasets, as well as cross-testing, demonstrated an improvement in the performance of EpitopeVec over other methods in terms of accuracy and area under the curve (AUC). As the predictive performance depended on the species origin of the respective antigens (viral, bacterial, eukaryotic), we also trained our method on a large viral dataset to create a dedicated linear viral BCE predictor with improved cross-testing performance.
AVAILABLITY: The software is available at https://github.com/hzi-bifo/epitope-prediction.
SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
© The Author(s) 2021. Published by Oxford University Press.
References
- J Biomed Inform. 2015 Feb;53:405-14 - PubMed
- Sci Rep. 2019 Mar 5;9(1):3577 - PubMed
- Vaccine. 1999 Sep;18(3-4):311-4 - PubMed
- NAR Genom Bioinform. 2020 Jun;2(2):lqaa015 - PubMed
- Proc Natl Acad Sci U S A. 1995 Sep 12;92(19):8700-4 - PubMed
- Nucleic Acids Res. 2000 Jan 1;28(1):45-8 - PubMed
- Int J Pept Res Ther. 2020;26(2):1155-1163 - PubMed
- PLoS One. 2013 May 07;8(5):e62216 - PubMed
- J Mol Recognit. 2003 Jan-Feb;16(1):20-2 - PubMed
- J Mol Biol. 1976 Jun 14;104(1):59-107 - PubMed
- Chem Rev. 2020 Mar 25;120(6):3210-3229 - PubMed
- Curr Top Med Chem. 2019;19(2):105-115 - PubMed
- Immunome Res. 2006 Apr 24;2:2 - PubMed
- Pac Symp Biocomput. 2002;:564-75 - PubMed
- Bioinformatics. 2017 Jan 1;33(1):42-48 - PubMed
- Protein Eng Des Sel. 2009 Mar;22(3):113-20 - PubMed
- Nat Biotechnol. 2011 May 15;29(7):644-52 - PubMed
- Proteins. 2020 Mar;88(3):397-413 - PubMed
- Proteins. 2006 Oct 1;65(1):40-8 - PubMed
- Curr Pharm Des. 2010;16(28):3149-57 - PubMed
- Methods Enzymol. 1991;203:176-201 - PubMed
- BioData Min. 2020 Apr 17;13:1 - PubMed
- J Immunol Res. 2017;2017:2680160 - PubMed
- PLoS One. 2013 Nov 11;8(11):e78605 - PubMed
- J Mol Recognit. 2008 Jul-Aug;21(4):243-55 - PubMed
- Rev Med Virol. 2009 Mar;19(2):77-96 - PubMed
- Bioinformatics. 2018 Sep 1;34(17):i773-i780 - PubMed
- Genome Biol. 2019 Nov 19;20(1):244 - PubMed
- Nat Biotechnol. 2015 Aug;33(8):831-8 - PubMed
- PLoS One. 2012;7(9):e45152 - PubMed
- BMC Genomics. 2005 May 29;6:79 - PubMed
- Nucleic Acids Res. 2010 Jan;38(Database issue):D854-62 - PubMed
- Bioinformatics. 2015 Sep 15;31(18):2939-46 - PubMed
- Bioinformatics. 2018 Jul 1;34(13):i32-i42 - PubMed
- Bioinformatics. 2021 May 1;37(4):448-455 - PubMed
- Methods Mol Biol. 2007;409:v-vi - PubMed
- Biol Direct. 2013 Oct 30;8:27 - PubMed
- BioData Min. 2015 Apr 29;8:14 - PubMed
- Nature. 1986 Aug 21-27;322(6081):747-8 - PubMed
- PLoS One. 2015 Nov 10;10(11):e0141287 - PubMed
- Infect Dis Poverty. 2020 Jul 10;9(1):88 - PubMed
- Comput Biol Chem. 2016 Jun;62:82-95 - PubMed
- Bioinformatics. 2008 Apr 1;24(7):924-31 - PubMed
- Nucleic Acids Res. 2017 Jul 3;45(W1):W24-W29 - PubMed
- Genome Biol. 2014 Mar 03;15(3):R46 - PubMed
- BioDrugs. 2010 Feb 1;24(1):1-8 - PubMed
- Cell. 2013 Jan 17;152(1-2):327-39 - PubMed
- Protein Sci. 2005 Jan;14(1):246-8 - PubMed
- Amino Acids. 2007 Sep;33(3):423-8 - PubMed
- J Virol. 1985 Sep;55(3):836-9 - PubMed
- Cell Host Microbe. 2020 Apr 8;27(4):671-680.e2 - PubMed
- Front Immunol. 2018 Jul 27;9:1695 - PubMed
- J Immunol Res. 2016;2016:6760830 - PubMed
- BMC Bioinformatics. 2013 Mar 09;14:90 - PubMed
- Int J Mol Sci. 2021 Mar 22;22(6): - PubMed
- Bioinformatics. 2010 Mar 1;26(5):680-2 - PubMed
- J Mol Graph. 1993 Sep;11(3):204-10, 191-2 - PubMed
- Sci Rep. 2018 Oct 8;8(1):14904 - PubMed
- FEBS Lett. 1990 Dec 10;276(1-2):172-4 - PubMed
- Curr Protein Pept Sci. 2003 Aug;4(4):299-308 - PubMed
Publication Types