A Web Service for Fragment-based Selectivity Analysis of Drug Leads

基于片段的先导药物选择性分析的 Web 服务

基本信息

批准号：
10701896
负责人：
John Laurence Kulp III
金额：
$ 96.65万
依托单位：
CONIFER POINT PHARMACEUTICALS, LLC
依托单位国家：
美国
项目类别：
财政年份：
2020
资助国家：
美国
起止时间：
2020-07-01 至 2025-08-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10701896
关键词：
2019-nCoV Acceleration Achievement Address Affinity Algorithms Benchmarking Big Data Binding Binding Sites Biological Sciences Biotechnology Capital Case Study Chemicals Chemistry Clinical Trials Cloud Computing Cloud Service Collection Coniferophyta Consumption DHFR gene Data Data Set Databases Doctor of Philosophy Drug Design Drug Targeting Evaluation Family Family member Feedback Free Energy Funding Generations Geometry Goals Grant Hand Homology Modeling Image Individual Internet Learning Libraries Maps Medicine Methodology Modification Mutation PTGS1 gene Patients Pattern Peptide Hydrolases Pharmaceutical Preparations Pharmacologic Substance Phase Phosphotransferases Phylogenetic Analysis Protein Family Protein Fragment Proteins Publications Publishing Research Personnel Resources Running SIRT1 gene Sales Sampling Scientist Services Sirtuins Small Business Innovation Research Grant Structure Techniques Technology Therapeutic Time Toxic effect Trees United States National Institutes of Health Validation Visualization Water Work candidate identification clinical candidate cloud platform commercialization computational chemistry cost design drug discovery empowerment enzyme pathway experience improved inhibitor innovation insight open source p38 Mitogen Activated Protein Kinase pre-clinical priority pathogen programs protein data bank protein protein interaction protein structure prototype rational design repository scale up simulation success therapeutic protein therapeutic target tool web app web interface web services

2019-nCoV Acceleration Achievement Address Affinity Algorithms Benchmarking Big Data Binding Binding Sites Biological Sciences Biotechnology Capital Case Study Chemicals Chemistry Clinical Trials Cloud Computing Cloud Service Collection Coniferophyta Consumption DHFR gene Data Data Set Databases Doctor of Philosophy Drug Design Drug Targeting Evaluation Family Family member Feedback Free Energy Funding Generations Geometry Goals Grant Hand Homology Modeling Image Individual Internet Learning Libraries Maps Medicine Methodology Modification Mutation PTGS1 gene Patients Pattern Peptide Hydrolases Pharmaceutical Preparations Pharmacologic Substance Phase Phosphotransferases Phylogenetic Analysis Protein Family Protein Fragment Proteins Publications Publishing Research Personnel Resources Running SIRT1 gene Sales Sampling Scientist Services Sirtuins Small Business Innovation Research Grant Structure Techniques Technology Therapeutic Time Toxic effect Trees United States National Institutes of Health Validation Visualization Water Work candidate identification clinical candidate cloud platform commercialization computational chemistry cost design drug discovery empowerment enzyme pathway experience improved inhibitor innovation insight open source p38 Mitogen Activated Protein Kinase pre-clinical priority pathogen programs protein data bank protein protein interaction protein structure prototype rational design repository scale up simulation success therapeutic protein therapeutic target tool web app web interface web services

展开

项目摘要

Significance: The goal of selective drug binding is a fundamental objective in the discovery and optimization of a compound on a trajectory toward helping patients and becoming an approved medication. To aid in this goal, our proposal aims to commercialize an innovative tool that addresses target selectivity in a rational de- sign methodology. The prototype tool leverages our large database of chemical fragment binding maps on therapeutically relevant proteins. These include over 100,000 maps covering over 600 drug targets including those on the NIH priority pathogen list, all SARS-CoV-2 structures, and almost 100 structures from the top life science venture capital firms. Searching spatial and energetic binding patterns of fragments gives valuable in- sights into designing selective or pan-selectivity in drugs. Conifer Point’s main product, BMaps, is supported by NIH SBIR grants and will be commercially released in 2022. The product has the largest repository of fragment binding data, affordable/accurate water molecule maps, and is integrated with other standard chemistry tools. To extract the information from the big data of fragment maps, a web service—backed by cloud computing—provides the data in a rational drug design appli- cation. Our prototype selectivity tool, BMaps-select, now allows users to identify candidate compounds by vis- ualizing how and why compounds interact with multiple target proteins, and by exploring suggested compound modifications derived from chemical fragment binding maps across multiple target proteins. The result is higher affinity and more selective compounds that specifically exploit the details of binding sites of a particular protein or protein family. BMaps and BMaps-select are low-cost, easy-to-learn, and available everywhere via the Web. Innovation: To date, no tools are available for the rational design of selectivity across 100s of proteins using fragment maps. Final compound evaluation can be done on individual proteins, but this is time consuming and inefficient. Our solution, BMaps-select, offers the potential for users to design across 100s of proteins within seconds and evaluate compounds across the same hundreds of proteins in minutes with easy-to-use tools. Approach: Our approach follows a similar trajectory to our prior work. First, we will pre-compute a large set of fragment maps (>1 million maps) for important therapeutic protein families. Build a web interface that can lev- erage the data and allow for selectivity design across hundreds of proteins. Lastly, we will validate the data and tools using open source and proprietary datasets, including a unique kinome-wide dataset of >600 inhibitors. Overall Impact: Drug selectivity is an important and fundamental obstacle in the progression of preclinical leads. BMaps-select offers the opportunity to be a first-in-class innovation to help accelerate preclinical drug discovery and to reduce toxicities due to off-target interactions, thus improving success rates of clinical trials.

意义：选择性药物结合的目标是发现和优化的基本目标轨迹上的化合物旨在帮助患者并成为批准的药物。为此提供帮助目标，我们的建议旨在将创新工具商业化，该工具以合理的方式解决目标选择性符号方法。原型工具利用了我们的大型化学片段结合图的数据库治疗相关蛋白质。其中包括超过100,000张地图，涵盖了600多个药物目标 NIH优先病原体列表，所有SARS-COV-2结构以及最高寿命的近100个结构科学风险投资公司。搜索碎片的空间和充满活力的结合模式可有价值视力在药物中设计选择性或泛选择性。针叶树点的主要产品BMAPS由NIH SBIR赠款支持，并将在商业上发布 2022年。该产品具有最大的碎片结合数据存储库，负担得起/准确的水分子地图，并与其他标准化学工具集成在一起。从大数据中提取信息片段图，一种由云计算返回的Web服务，以合理的药物设计应用程序提供了数据阳离子。我们的原型选择性工具BMAPS选择，现在允许用户通过VIS-识别候选化合物化合物如何以及为什么与多种靶蛋白相互作用，并通过探索建议的化合物从多个靶蛋白跨化学片段结合图得出的修饰。结果更高亲和力和更多选择性化合物，这些化合物专门利用特定蛋白质的结合位点的细节或蛋白质家族。 Bmaps和Bmaps选择是低成本，易于学习的，并且可以通过网络在任何地方使用。创新：迄今碎片地图。最终的复合评估可以对单个蛋白质进行，但这很耗时，效率低下。我们的解决方案BMAPS选择，为用户提供了在100张蛋白质中设计的潜力秒并使用易于使用的工具在几分钟内评估相同数百种蛋白质的化合物。方法：我们的方法遵循与我们先前的工作相似的轨迹。首先，我们将预先计算大量重要的治疗蛋白家族的片段图（> 100万张图）。构建一个可以利用的Web界面掌握数据并允许数百种蛋白质的选择性设计。最后，我们将验证数据，使用开源和专有数据集的工具，包括> 600个抑制剂的独特范围内数据集。总体影响：药物选择性是临床前的重要障碍铅。 Bmaps选择提供了成为一类创新的机会，以帮助加速临床前药物发现并减少因脱靶相互作用而导致的毒性，从而提高了临床试验的成功率。