SBIR Phase I: AI-assisted identification of small molecules for targeted repair of vascular barrier dysfunctions

小型企业创新研究(SBIR)第一阶段项目的更广泛影响/商业潜力是确定血管渗漏疾病的治疗方法。失控的血管渗漏障碍是一种常见的病理过程，可导致跨越多个器官的组织损伤和60多种临床情况。目前，还没有解决血管泄漏的药物疗法。现有的解决方案侧重于提供支持性护理或减少炎症，而没有解决潜在的机制。该项目提出了一种修复血管渗漏的新方法，作为一种治疗干预。通过建立第一个用于识别修复血管渗漏的小分子的药物发现工作流程，该项目将使多种情况下的药物流水线得以开发。第一个目标将是急性呼吸窘迫综合征(ARDS)，它占重症监护病房(ICU)入院人数的10%，是ICU死亡的主要原因。在全球范围内，它每年影响300多万患者。拟议的解决方案将减少死亡人数和ICU的成本。这个小企业创新研究(SBIR)第一阶段项目寻求验证一种新的基于结构的药物筛选平台，该平台旨在识别激活负责修复血管渗漏的分子通路的小分子。建议的平台包括用于配体结合位点预测的新型机器学习方法的独特组合，能够在几分钟内(5,000种化合物/秒)筛选具有定向吸收、分布、代谢和排泄毒性(ADME-Tox)特征的超大(超过10亿分子)化合物文库的快速对接算法。人工智能引导的对接方法与体外高通量分析相结合，测量人体组织生理相关微环境中的血管泄漏机制，以选择针对血管泄漏障碍的候选对象。在这个项目中，这个分级工作流程的步骤将被验证并应用于第一个目标，从而确定一组新的候选药物。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势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 small Business Innovation Research (SBIR) Phase I project is the identification of treatments for vascular leak disorders. Uncontrolled vascular leak disorders are common pathological processes that lead to tissue damage across multiple organs and over 60 clinical conditions. Currently, there are no drug-based therapies that address vascular leaks. Available solutions are focused on providing supportive care or decreasing inflammation, without addressing the underlying mechanism. This project is proposing a new approach to repair vascular leaks as a therapeutic intervention. By establishing the first drug discovery workflow for the identification of small molecules that repair vascular leakage, this project will enable the development of a pipeline of drugs for multiple conditions. The first condition targeted will be acute respiratory distress syndrome (ARDS), which accounts for 10% of intensive care unit (ICU) admissions and is the leading cause of mortality in ICU. Globally, it affects more than 3 million patients yearly. The proposed solution will decrease the number of deaths and the costs for ICU.This Small Business Innovation Research (SBIR) Phase I project seeks to validate a new structure-based drug screening platform designed to identify small molecules that activate the molecular pathways responsible for repairing vascular leakage. The proposed platform consists of a unique combination of novel machine learning methods for ligand-binding site prediction, fast docking algorithm capable of screening ultra-large (over a billion molecule) compound libraries with targeted absorption, distribution, metabolism, and excretion-toxicity (ADME-Tox) profile within minutes (5,000 compounds/second). The AI-guided docking approach is combined with in vitro high-throughput assays measuring the mechanisms of vascular leak in a physiologically relevant microenvironment of human tissues to select candidates targeting vascular leak disorders. In this project, the steps of this tiered workflow will be validated and applied to the first target, leading to the identification of a set of new drug candidates.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.