Synergistic integration of topology and machine learning for the predictions of protein-ligand binding affinities and mutation impacts

拓扑和机器学习的协同集成，用于预测蛋白质-配体结合亲和力和突变影响

• 批准号: 10189006
• 负责人: Guowei Wei
• 金额: $ 11.64万
• 依托单位: MICHIGAN STATE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-08-01 至 2022-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10189006
• 关键词: 2019-nCoV; Address; Adopted; Affinity; Antibodies; Antibody Affinity; Artificial Intelligence; Attention; Award; Binding; Binding Proteins; COVID-19; COVID-19 vaccine; Cells; Common Cold; Conflict (Psychology); Coronavirus; Crystallization; Development; Diagnostic; Diagnostic Reagent; Diagnostic tests; Drug resistance; Drug usage; Evolution; Failure; Future; Genetic Transcription; Genome; Goals; Grant; Heterogeneity; Induced Mutation; Influenza A virus; Machine Learning; Manuscripts; Mathematics; Medicine; Methods; Mission; Modeling; Modification; Molecular; Mutate; Mutation; National Institute of General Medical Sciences; Nucleotides; Peptidyl-Dipeptidase A; Pharmaceutical Preparations; Phase; Preventive measure; Preventive vaccine; Process; Proteins; RNA Viruses; Recurrence; Reporting; Research; Rhinovirus; SARS coronavirus; Severe Acute Respiratory Syndrome; Site; Structure; Therapeutic; Therapeutic antibodies; Time; Vaccines; Viral; Virulence; Virus; Work; antigen antibody binding; autoencoder; base; combat; design; experience; flu; improved; inhibitor/antagonist; news; pandemic disease; protein folding; receptor; receptor binding; reconstruction; resistance mutation; success; vaccine development

Project Summary The success of the ongoing battle with coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) depends crucially on the availability of effective diagnostics, vaccines, antibody therapeutics, and small-molecular drugs. Although SARS-CoV-2 mutates slower than the viruses that cause the flu and the common cold, it has had more than 8300 observed single mutations on its genome of 29,900 nucleotides by June 1, 2020. We show that these mutations might have devastating effects on COVID-19 diagnostics, vaccines, antibody therapeutics, and small-molecular drugs (J. Chem. Inf. Model. In press). We will develop new artificial intelligence (AI) to forecast SARS-CoV-2 future mutations. Leveraging on state-of-art methods developed under the present R01 award, we will design mutation- resistant vaccines, antibody therapeutics, and small-molecular drugs. The CPUs and GPUs requested in this supplement will be essential for my lab to continue the research of the present R01 award and to apply the methods developed in this award to attack fundamental problems in combating COVID-19.

项目摘要 2019年冠状病毒病（COVID-19）是由严重的 急性呼吸道综合征冠状病毒2型（SARS-CoV-2）在很大程度上取决于 有效的诊断方法、疫苗、抗体治疗方法和小分子药物。虽然 SARS-CoV-2的变异速度比引起流感和普通感冒的病毒慢， 到2020年6月1日，在其29，900个核苷酸的基因组上观察到超过8300个单突变。 我们发现这些突变可能对COVID-19诊断产生破坏性影响， 疫苗、抗体治疗剂和小分子药物（J. Chem. Inf. Model.待印）。我们 将开发新的人工智能（AI）来预测SARS-CoV-2未来的突变。利用 根据目前R 01奖开发的最先进的方法，我们将设计突变- 耐药疫苗、抗体治疗剂和小分子药物。CPU和GPU 在本补充中所要求的将是必不可少的，我的实验室继续目前的研究 R 01奖项，并应用该奖项中开发的方法来解决以下基本问题： 抗击COVID-19