AB-Gen:Antibody Library Design with Generative Pre-trained Transformer and Deep Reinforcement Learning
作者机构：Computational Bioscience Research Center(CBRC)King Abdullah University of Science and TechnologyThuwal 23955-6900Saudi Arabia State Key Laboratory of Structural ChemistryFujian Institute of Research on the Structure of MatterChinese Academy of SciencesFuzhou 350002China University of Chinese Academy of SciencesBeijing 100049China Syneron TechnologyGuangzhou 510000China Fujian Provincial Key Laboratory of Theoretical and Computational ChemistryFuzhou 361005China
出 版 物：《Genomics, Proteomics & Bioinformatics》 (基因组蛋白质组与生物信息学报（英文版）)
年 卷 期：2023年第21卷第5期
页 面：1043-1053页
核心收录：
学科分类：0710[理学-生物学] 12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 081104[工学-模式识别与智能系统] 0835[工学-软件工程] 0811[工学-控制科学与工程] 081202[工学-计算机软件与理论] 0812[工学-计算机科学与技术（可授工学、理学学位）]
基 金：supported in part by the Office of Research Administration(ORA),King Abdullah University of Science and Technology(KAUST),Saudi Arabia(Grant Nos.FCC/1/1976-44-01,FCC/1/1976-45-01,REI/1/5234-01-01,and URF/1/4352-01-01) the National Natural Science Foundation of China(Grant No.22273107)
主 题：Protein design Transformer Reinforcement learning Generative modeling Multi-objective optimization
摘 要：Antibody leads must fulfill multiple desirable properties to be clinical *** due to the low throughput in the experimental procedure,the need for such multiproperty optimization causes the bottleneck in preclinical antibody discovery and development,because addressing one issue usually causes *** developed a reinforcement learning(RL)method,named AB-Gen,for antibody library design using a generative pre-trained transformer(GPT)as the policy network of the RL *** showed that this model can learn the antibody space of heavy chain complementarity determining region 3(CDRH3)and generate sequences with similar property ***,when using human epidermal growth factor receptor-2(HER2)as the target,the agent model of AB-Gen was able to generate novel CDRH3 sequences that fulfill multi-property ***,509 generated sequences were able to pass all property filters,and three highly conserved residues were *** importance of these residues was further demonstrated by molecular dynamics simulations,consolidating that the agent model was capable of grasping important information in this complex optimization ***,the ABGen method is able to design novel antibody sequences with an improved success rate than the traditional propose-then-filter *** has the potential to be used in practical antibody design,thus empowering the antibody discovery and development *** source code of AB-Gen is freely available at Zenodo(https://***/10.5281/zenodo.7657016)and BioCode(https://***/biocode/tools/BT007341).
维普期刊数据库