STTR Phase I: COVID-19: AI-based Development of Neutralizing Antibodies for SARS-CoV-2

该STTR项目的更广泛影响/商业潜力将导致工程抗体的开发，可用于为感染COVID-19的患者提供被动免疫和治疗。这些中和抗体也可作为COVID-19感染高风险人群的预防措施。与人体自然产生的抗体相比，这种工程抗体提供了更广泛的潜在解决方案，有可能提供更有效的解决方案。高通量筛选、下一代测序和基于人工智能的抗体设计相结合，可以系统地探索广泛的抗体序列。该平台技术将对未来新型冠状病毒或现有冠状病毒突变形式的爆发做出高度反应。这项技术将成为一种平台技术，对冠状病毒以外的其他疾病的治疗方法也很有用。由于目前正在发生的COVID-19大流行，这一领域的解决方案具有高度相关性。该STTR一期项目提出通过提供所需的抗体序列突变结合数据，将基于人工智能的抗体工程提升到一个新的水平，从而极大地实现人工智能和机器学习抗体工程方法。目前可用的抗体数据集有数千个数据点，而该项目提出要生成数千万个数据点的数据集。该项目还将生成阳性和阴性抗体结合数据，从而产生更高性能的学习抗体结合模型。该项目允许测试一个假设，即合成抗体可以等于或优于自然产生的抗体，以中和SARS-CoV-2的传染性。自然界在产生抗体方面有自己的一套规则和限制，而这项提议的方法可能会开发出更广泛的抗体变体。本研究将集中于通过酵母展示、高通量FACS分选和下一代测序相结合，发现一些靶向SARS-CoV-2刺突蛋白受体结合域（RBD）的高亲和力抗体。将这些高通量数据生成工作流程与最新的深度神经网络相结合，将产生一种新的方法，可以快速有效地发现高效抗体，既适用于当前的大流行，也适用于可能随之而来的其他大流行。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this STTR project will lead to the development of engineered antibodies for that can be used to provide passive immunity and treatment to patients infected with COVID-19. These neutralizing antibodies can also be administered as preventative measures for populations at high risk of contracting COVID-19. Such engineered antibodies present a wider range of potential solutions than those produced naturally in the human body, potentially allowing more effective solutions. The proposed combination of high-throughput screening, next-generation-sequencing and AI-based antibody design allows systematic exploration of vast ranges of antibody sequences. This platform technology will be highly responsive to future outbreaks of novel coronaviruses or mutated forms of existing coronaviruses. The technology will be a platform technology which is would be useful going forward for other therapeutics for different diseases beyond coronavirus. Solutions in this space are highly relevant due to the current ongoing COVID-19 pandemic.This STTR Phase I project proposes to greatly enable AI and machine learning antibody engineering approaches by providing the needed antibody sequence mutation binding data that will take AI-based antibody engineering to a new level. Currently available antibody datasets number in the thousands of datapoints and this project proposes to generate datasets that number in the tens of millions. The project will also be generating both positive and negative antibody binding data, leading to higher performing learned antibody binding models. This project allows testing the hypothesis that synthetic antibodies can be the equal of, or better than, naturally occurring antibodies for neutralizing SARS-CoV-2 infectivity. Nature has its own set of rules and limitations for generating antibodies and the propsoals' approach could potentially develop a much wider range of antibody variations. This work will be laser-focused on discovering a number of high-affinity antibodies targeting the receptor binding domain (RBD) of the SARS-CoV-2 spike protein through the combination of yeast-display, high-throughput FACS sorting and next-generation-sequencing. Combining these high-throughput data generation workflows with the latest deep neural networks will lead to a new methodology that can quickly and efficiently discover high performing antibodies, both for the current pandemic and others that may follow.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.