Generative neural networks for structure-based antibody design

用于基于结构的抗体设计的生成神经网络

• 批准号: 10705666
• 负责人: Possu Huang
• 金额: $ 32.95万
• 依托单位: STANFORD UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-17 至 2027-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10705666
• 关键词: 2019-nCoV; 3-Dimensional; APLN gene; Acceleration; Address; Adopted; Agonist; Algorithm Design; Antibodies; Antivenins; Architecture; Area; Binding; Biotechnology; CXCR4 gene; Cardiac Myocytes; Clinical; Complex; Computational algorithm; Computing Methodologies; Data; Data Science; Development; Diagnostic tests; Docking; Engineering; Environment; Epitopes; Evolution; Experimental Designs; Face; Fingers; G-Protein-Coupled Receptors; Generations; Goals; Home; Image; Immune; Immunoglobulins; Learning; Libraries; Marketing; Medical; Medical Technology; Methods; Modeling; Modernization; Molecular; Molecular Conformation; Myocardial Ischemia; Neural Network Simulation; Outcome; Performance; Play; Positioning Attribute; Probability; Process; Protein Engineering; Proteins; Recording of previous events; Reporting; Research; Research Personnel; Role; Sampling; Scheme; Side; Signal Transduction; Snake Venoms; Snakes; Specificity; Speed; Structure; Technology; Testing; Therapeutic; Time; Toxin; Training; Variant; Vertebral column; antibody engineering; artificial neural network; chemokine receptor; cluster computing; combinatorial; computational platform; cost; deep learning; deep neural network; design; detection platform; flexibility; generative adversarial network; improved; interest; interfacial; knowledge base; loss of function; method development; model design; molecular recognition; nanobodies; neural network; novel; programs; protein structure; receptor; receptor binding; response; scaffold; screening; tool

PROGRAM SUMMARY/ABSTRACT As a molecular detection platform, antibodies have growing importance in modern medical technology, ranging from diagnostic tests, to imaging, to therapeutics. The current market size for antibodies and their related products is estimated to be around $200 billion USD. The growing need for antibodies with customized specificity provides a rich environment for engineering efforts. Computational protein design has seen rapid progress in recent years. Many methods have been developed to address antibody engineering needs. Researchers have hoped that, through modeling and design, the cost for antibody development and improvements can be reduced and the pace for creating new targeting molecules can be expedited. In recent years, the experimental pipeline has been streamlined, but even so, extensive libraries and screen campaigns are usually required to get an initial binding signal. A major advancement would be to directly design a binder from scratch, providing a signal for potential optimization by artificial evolution. Current computational methods, however, have not taken a leading role due to a number of shortcomings with the current modeling approach. We have extensive expertise in protein design and have pioneered the use of generative neural network models for protein structures in recent years. We have observed several key advantages in neural network approaches over existing methods: namely, their ability to make inferences, interpolate, incorporate topological information, and accelerate sampling. These advantages can be developed independently or used in conjunction with existing methods, and they can significantly boost the performance of protein design. This project aims at leveraging several new advances we have developed to date to inspire new strategies in response to the challenges in antibody engineering, or AI-based protein design in general. We will develop new tools and design pipelines for expanding the specificities for multi- specific antibodies and customizing epitope-specific antibodies (using snake venoms and CXCR4 as targets). This project will deliver both computational methods and constructs that can be deployed in clinical settings. The results from this research will be highly impactful.

程序摘要/摘要 作为一个分子检测平台，抗体在现代医学中越来越重要 从诊断测试到成像到治疗药的技术。当前的市场规模 抗体及其相关产品估计约为2000亿美元。成长 具有定制特异性的抗体的需求为工程工作提供了丰富的环境。 近年来，计算蛋白设计取得了迅速的进步。许多方法已经 开发以满足抗体工程需求。研究人员希望通过 建模和设计，可以降低抗体开发和改进的成本，并 可以加快创建新的靶向分子的速度。近年来，实验 管道已经简化，但即使如此，广泛的图书馆和屏幕活动也是如此 需要获得初始绑定信号。主要进步是直接设计活页夹 从头开始，通过人工进化提供信号以进行潜在优化。当前的 但是，由于许多缺点，计算方法没有发挥主导作用 采用当前的建模方法。我们在蛋白质设计方面拥有广泛的专业知识，并具有 近年来，率先将生成神经网络模型用于蛋白质结构。我们 在神经网络方法中观察到了几个关键优势，而不是现有方法： 也就是说，他们做出推断，插入，结合拓扑信息的能力以及 加速采样。这些优点可以独立开发或结合使用 使用现有方法，它们可以显着提高蛋白质设计的性能。这 项目旨在利用我们迄今为止已经开发的几个新进步来激发新的 应对抗体工程挑战或基于AI的蛋白质设计的策略 一般的。我们将开发新的工具和设计管道，以扩大多种多样的特异性 特定抗体和定制表位特异性抗体（使用蛇毒和CXCR4 作为目标）。该项目将提供可以是计算方法和构造 部署在临床环境中。这项研究的结果将具有很高的影响力。