Abstract: Antibodies currently comprise the predominant treatment modality for a variety of diseases; therefore, optimizing their properties rapidly and efficiently is an indispensable step in antibody-based drug development. Inspired by the great success of artificial intelligence-based algorithms, especially deep learning-based methods in the field of biology, various computational methods have been introduced into antibody optimization to reduce costs and increase the success rate of lead candidate generation and optimization. Herein, we briefly review recent progress in deep learning-based antibody optimization, focusing on the available datasets and algorithm input data types that are crucial for constructing appropriate deep learning models. Furthermore, we discuss the current challenges and potential solutions for the future development of general-purpose deep learning algorithms in antibody optimization.

基于深度学习方法的抗体优化研究进展

[1]AbramsonJ, AdlerJ, DungerJ, et al., 2024. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature, 630(8016):493-500.

[2]AkbarR, BashourH, RawatP, et al., 2022. Progress and challenges for the machine learning-based design of fit-for-purpose monoclonal antibodies. mAbs, 14(1):2008790.

[3]AkterJ, KhouryDS, AogoR, et al., 2019. Plasmodium-specific antibodies block in vivo parasite growth without clearing infected red blood cells. PLoS Pathog, 15(2):e1007599.

[4]AnsariHR, FlowerDR, RaghavaGPS, 2010. AntigenDB: an immunoinformatics database of pathogen antigens. Nucleic Acids Res, 38(suppl_1):D847-D853.

[5]ApweilerR, BairochA, WuCH, et al., 2004. UniProt: the Universal Protein knowledgebase. Nucleic Acids Res, 32(suppl_1):D115-D119.

[6]BairochA, ApweilerR, 1997. The SWISS-PROT protein sequence data bank and its supplement TrEMBL. Nucleic Acids Res, 25(1):31-36.

[7]BaldoBA, 2022. Immune- and non-immune-mediated adverse effects of monoclonal antibody therapy: a survey of 110 approved antibodies. Antibodies, 11(1):17.

[8]BennettNR, WatsonJL, RagotteRJ, et al., 2024. Atomically accurate de novo design of single-domain antibodies. bioRxiv, preprint.

[9]BermanHM, WestbrookJ, FengZK, et al., 2000. The protein data bank. Nucleic Acids Res, 28(1):235-242.

[10]BonicheC, RossiSA, KischkelB, et al., 2020. Immunotherapy against systemic fungal infections based on monoclonal antibodies. J Fungi, 6(1):31.

[11]ChenJH, WeiGW, 2022. Mathematical artificial intelligence design of mutation-proof COVID-19 monoclonal antibodies. Commun Inf Syst, 22(3):339-361.

[12]ChenJM, SawyerN, ReganL, 2013. Protein‒protein interactions: general trends in the relationship between binding affinity and interfacial buried surface area. Protein Sci, 22(4):510-515.

[13]CrescioliS, KaplonH, ChenowethA, et al., 2024. Antibodies to watch in 2024. mAbs, 16(1):2297450.

[14]do PazoC, NawazK, WebsterRM, 2021. The oncology market for antibody‒drug conjugates. Nat Rev Drug Discov, 20(8):583-584.

[15]DoronI, MeskoM, LiXV, et al., 2021. Mycobiota-induced IgA antibodies regulate fungal commensalism in the gut and are dysregulated in Crohn’s disease. Nat Microbiol, 6(12):1493-1504.

[16]DouguetD, ChenHC, TovchigrechkoA, et al., 2006. Dockground resource for studying protein‒protein interfaces. Bioinformatics, 22(21):2612-2618.

[17]DunbarJ, KrawczykK, LeemJ, et al., 2014. SAbDab: the structural antibody database. Nucl Acids Res, 42(D1):D1140-D1146.

[18]EvansR, O'NeillM, PritzelA, et al., 2021. Protein complex prediction with AlphaFold-Multimer. bioRxiv, preprint.

[19]GravesJ, ByerlyJ, PriegoE, et al., 2020. A review of deep learning methods for antibodies. Antibodies, 9(2):12.

[20]GuoZL, YamaguchiR, 2022. Machine learning methods for protein-protein binding affinity prediction in protein design. Front Bioinform, 2:1065703.

[21]GuptaR, SrivastavaD, SahuM, et al., 2021. Artificial intelligence to deep learning: machine intelligence approach for drug discovery. Mol Divers, 25(3):1315-1360.

[22]HernandezI, BottSW, PatelAS, et al., 2018. Pricing of monoclonal antibody therapies: higher if used for cancer? Am J Manag Care, 24(2):109-112.

[23]HouWP, ShangXY, JiZC, 2023. Benchmarking large language models for genomic knowledge with GeneTuring. bioRxiv, preprint.

[24]JankauskaitėJ, Jiménez-GarcíaB, DapkūnasJ, et al., 2019. SKEMPI 2.0: an updated benchmark of changes in protein‒protein binding energy, kinetics and thermodynamics upon mutation. Bioinformatics, 35(3):462-469.

[25]JinRF, YeQ, WangJK, et al., 2024. AttABseq: an attention-based deep learning prediction method for antigen‒antibody binding affinity changes based on protein sequences. Brief Bioinform, 25(4):bbae304.

[26]JumperJ, EvansR, PritzelA, et al., 2021. Highly accurate protein structure prediction with AlphaFold. Nature, 596(7873):583-589.

[27]KangY, LengDW, GuoJJ, et al., 2022. Sequence-based deep learning antibody design for in silico antibody affinity maturation. arXiv:2103.03724.

[28]KaplonH, CrescioliS, ChenowethA, et al., 2023. Antibodies to watch in 2023. mAbs, 15(1):2153410.

[29]KöhlerG, MilsteinC, 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature, 256(5517):495-497.

[30]LamHYI, OngXE, MutwilM, 2024. Large language models in plant biology. Trends Plant Sci, 29(10):1145-1155.

[31]LeeM, 2023. Recent advances in deep learning for protein-protein interaction analysis: a comprehensive review. Molecules, 28(13):5169.

[32]ListovD, GoverdeCA, CorreiaBE, et al., 2024. Opportunities and challenges in design and optimization of protein function. Nat Rev Mol Cell Biol, 25(8):639-653.

[33]LiuG, ZengHY, MuellerJ, et al., 2020. Antibody complementarity determining region design using high-capacity machine learning. Bioinformatics, 36(7):2126-2133.

[34]LiuXG, LuoYN, LiPY, et al., 2021. Deep geometric representations for modeling effects of mutations on protein-protein binding affinity. PLoS Comput Biol, 17(8):e1009284.

[35]LubianaT, LopesR, MedeirosP, et al., 2023. Ten quick tips for harnessing the power of ChatGPT in computational biology. PLoS Comput Biol, 19(8):e1011319.

[36]MakowskiEK, KinnunenPC, HuangJ, et al., 2022. Co-optimization of therapeutic antibody affinity and specificity using machine learning models that generalize to novel mutational space. Nat Commun, 13:3788.

[37]MasonDM, FriedensohnS, WeberCR, et al., 2021. Optimization of therapeutic antibodies by predicting antigen specificity from antibody sequence via deep learning. Nat Biomed Eng, 5(6):600-612.

[38]MastropietroA, PasculliG, BajorathJ, 2023. Learning characteristics of graph neural networks predicting protein‒ligand affinities. Nat Mach Intell, 5(12):1427-1436.

[39]MeliR, MorrisGM, BigginPC, 2022. Scoring functions for protein-ligand binding affinity prediction using structure-based deep learning: a review. Front Bioinform, 2:885983.

[40]MullardA, 2021. FDA approves 100th monoclonal antibody product. Nat Rev Drug Discov, 20(7):491-495.

[41]NotinP, RollinsN, GalY, et al., 2024. Machine learning for functional protein design. Nat Biotechnol, 42(2):216-228.

[42]O'SheaK, NashR, 2015. An introduction to convolutional neural networks. arXiv:1511.08458.

[43]PiresDEV, AscherDB, 2016. mCSM-AB: a web server for predicting antibody‒antigen affinity changes upon mutation with graph-based signatures. Nucleic Acids Res, 44(W1):W469-W473.

[44]RéauM, RenaudN, XueLC, et al., 2023. DeepRank-GNN: a graph neural network framework to learn patterns in protein‒protein interfaces. Bioinformatics, 39(1):btac759.

[45]ShanSS, LuoST, YangZQ, et al., 2022. Deep learning guided optimization of human antibody against SARS-CoV-2 variants with broad neutralization. Proc Natl Acad Sci USA, 119(11):e2122954119.

[46]SilverD, HuangA, MaddisonCJ, et al., 2016. Mastering the game of Go with deep neural networks and tree search. Nature, 529(7587):484-489.

[47]SirinS, ApgarJR, BennettEM, et al., 2016. AB‐bind: antibody binding mutational database for computational affinity predictions. Protein Sci, 25(2):393-409.

[48]SoleymaniF, PaquetE, ViktorH, et al., 2022. Protein‒protein interaction prediction with deep learning: a comprehensive review. Comput Struct Biotechnol J, 20:5316-5341.

[49]SoleymaniF, PaquetE, ViktorHL, et al., 2023. ProtInteract: a deep learning framework for predicting protein‒protein interactions. Comput Struct Biotechnol J, 21:1324-1348.

[50]SunXY, YiCY, ZhuYF, et al., 2022. Neutralization mechanism of a human antibody with pan-coronavirus reactivity including SARS-CoV-2. Nat Microbiol, 7(7):1063-1074.

[51]SzklarczykD, KirschR, KoutrouliM, et al., 2023. The STRING database in 2023: protein‒protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res, 51(D1):D638-D646.

[52]ThirumalaiD, Visaga AmbiS, Vieira-PiresRS, et al., 2019. Chicken egg yolk antibody (IgY) as diagnostics and therapeutics in parasitic infections‒a review. Int J Biol Macromol, 136:755-763.

[53]ToselandCP, ClaytonDJ, McSparronH, et al., 2005. AntiJen: a quantitative immunology database integrating functional, thermodynamic, kinetic, biophysical, and cellular data. Immunome Res, 1:4.

[54]TovchigrechkoA, VakserIA, 2006. GRAMM-X public web server for protein‒protein docking. Nucleic Acids Res, 34(suppl_2):W310-W314.

[55]VitaR, MahajanS, OvertonJA, et al., 2019. The Immune Epitope Database (IEDB): 2018 update. Nucleic Acids Res, 47(D1):D339-D343.

[56]WangGY, LiuXH, WangK, et al., 2023. Deep-learning-enabled protein‒protein interaction analysis for prediction of SARS-CoV-2 infectivity and variant evolution. Nat Med, 29(8):2007-2018.

[57]WangML, CangZX, WeiGW, 2020. A topology-based network tree for the prediction of protein‒protein binding affinity changes following mutation. Nat Mach Intell, 2(2):116-123.

[58]WangRX, FangXL, LuYP, et al., 2004. The PDBbind database: collection of binding affinities for protein‒ligand complexes with known three-dimensional structures. J Med Chem, 47(12):2977-2980.

[59]WilmanW, WróbelS, BielskaW, et al., 2022. Machine-designed biotherapeutics: opportunities, feasibility and advantages of deep learning in computational antibody discovery. Brief Bioinform, 23(4):bbac267.

[60]WuZH, PanSR, ChenFW, et al., 2021. A comprehensive survey on graph neural networks. IEEE Trans Neural Netw Learn Syst, 32(1):4-24.

[61]YoungC, LauAWY, BurnettDL, 2022. B cells in the balance: offsetting self-reactivity avoidance with protection against foreign. Front Immunol, 13:951385.

[62]ZhaoY, HeB, XuF, et al., 2023. DeepAIR: a deep learning framework for effective integration of sequence and 3D structure to enable adaptive immune receptor analysis. Sci Adv, 9(32):eabo5128.

[63]ZhengW, ZhaoWJ, WuM, et al., 2020. Microbiota-targeted maternal antibodies protect neonates from enteric infection. Nature, 577(7791):543-548.

[64]ZurawskiDV, McLendonMK, 2020. Monoclonal antibodies as an antibacterial approach against bacterial pathogens. Antibiotics, 9(4):155.