NOVEL PARADIGMS FOR DRUG DISCOVERY: COMPUTATIONAL MULTITARGET SCREENING

药物发现的新范式：计算多目标筛选

• 批准号: 7979181
• 负责人: RAM SAMUDRALA
• 金额: $ 83万
• 依托单位: UNIVERSITY OF WASHINGTON
• 依托单位国家: 美国
• 财政年份: 2010
• 资助国家: 美国
• 起止时间: 2010-09-30 至 2015-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7979181
• 关键词: Adverse effects; Area; Award; Clinical; Communicable Diseases; Communities; Computational Biology; Dengue; Disease; Drug Delivery Systems; Effectiveness; Ensure; Excretory function; Family; Funding; Goals; Herpesviridae; Human; Immunocompromised Host; In Vitro; Infection; Lead; Machine Learning; Malaria; Metabolism; Methods; Pathway interactions; Pharmacologic Substance; Proteins; Protocols documentation; Risk; Science; Screening procedure; Source; Techniques; Technology; Therapeutic; Therapeutic Uses; Time; Variant; Virus; absorption; abstracting; base; cost; drug candidate; drug discovery; in vivo; inhibitor/antagonist; mortality; novel; pathogen; pre-clinical; prospective; small molecule; success; three dimensional structure

DESCRIPTION Abstract: We will create a comprehensive computational drug discovery platform by enhancing a novel technique for a dynamic, fragment based, screening of small molecule compounds against the three dimensional structures of multiple protein targets from infectious disease causing pathogens, followed by prospective in vitro and in vivo experimental verification. We will further modify the most promising lead candidates computationally and screen them against all known human proteins and variants simultaneously to assess for side effects against essential proteins, and to ensure that they possess safe and effective absorption, distribution, metabolism, and excretion profiles against major proteins in known drug delivery pathways. The top ranking leads will again be experimentally verified, and the computational protocol will be itera- tively refined using machine learning techniques. We will initially focus on discovering preclinical drug candidates against infections caused by all eight human herpes viruses (HHVs). This virus family infects billions of humans worldwide every year and is the source of significant mortality in immunocompromised patients. Broad spectrum therapeutics against these key pathogens will benefit the entire global community. In contrast to other computational efforts, my group has successfuly applied and experimentally verified their predictions of inhibitors to treat herpes, malaria, and dengue. This was accomplished at a fraction of the time, effort, and cost typically required by pharmaceutical companies. Our significant successes thus far attest to the efficacy of our drug discovery technologies. The Pioneer Award funds will therefore allow us to bridge the gap of discovering computationally predicted lead compounds and demonstrating their preclinical effectiveness for further clinical and therapeutic use. The ultimate goal is to create a comprehensive computational drug discovery pipeline, applicable to any disease, thereby increasing the suc

描述 摘要： 我们将创建一个全面的计算药物发现平台，通过增强一种新的技术，用于动态的，基于片段的，针对引起感染性疾病的病原体的多个蛋白质靶点的三维结构筛选小分子化合物，然后进行前瞻性的体外和体内实验验证。我们将通过计算进一步修改最有希望的先导候选物，并同时对所有已知的人类蛋白质和变体进行筛选，以评估对必需蛋白质的副作用，并确保它们对已知药物递送途径中的主要蛋白质具有安全有效的吸收，分布，代谢和排泄特征。排名靠前的线索将再次通过实验验证，计算方案将使用机器学习技术进行迭代优化。我们最初将专注于发现针对所有八种人类疱疹病毒（HHV）引起的感染的临床前候选药物。该病毒家族每年感染全球数十亿人，是免疫功能低下患者的重大死亡原因。针对这些关键病原体的广谱疗法将使整个全球社会受益。与其他计算工作相比，我的团队成功地应用并实验验证了他们对治疗疱疹，疟疾和登革热的抑制剂的预测。这是在一小部分的时间，精力和成本通常所需的制药公司。我们迄今取得的重大成功证明了我们药物发现技术的有效性。因此，先锋奖基金将使我们能够弥合发现计算预测的先导化合物和证明其临床前有效性以进一步临床和治疗用途的差距。最终目标是创建一个全面的计算药物发现管道，适用于任何疾病，从而增加药物的有效性。