A translational bioinformatics approach to elucidate and mitigate polypharmacy induced adverse drug reactions

阐明和减轻复方用药引起的药物不良反应的转化生物信息学方法

• 批准号: 10664024
• 负责人: Zackary Michael Falls
• 金额: $ 20.93万
• 依托单位: STATE UNIVERSITY OF NEW YORK AT BUFFALO
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-08-01 至 2027-07-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10664024
• 关键词: Achievement; Affinity; Age; Aptitude; Architecture; Area; Basic Science; Benzodiazepines; Binding; Binding Proteins; Bioinformatics; Biological; Biological Assay; Buprenorphine; Cause of Death; Cessation of life; Chemical Models; Chronic Disease; Clinical; Clinical Data; Clinical Decision Support Systems; Clinical Informatics; Clinical Skills; Complement; Computer Analysis; Computer software; Counseling; Creativeness; Data; Data Set; Databases; Development; Diagnosis; Docking; Drug Combinations; Drug Interactions; Drug Prescriptions; Drug usage; Electronic Health Record; Environment; Ethnic Origin; Event; FDA approved; Financial Hardship; Funding; Gender; Goals; Grant; Graph; Health; Healthcare Systems; High Prevalence; Human; Impaired cognition; In Vitro; Incidence; K-Series Research Career Programs; Knowledge; Knowledge acquisition; Medicine; Mentors; Methadone; Methodology; Methods; Naltrexone; Opiate Addiction; Opioid; Outcome; Overdose; Overdose reduction; Pathway interactions; Patients; Performance; Pharmaceutical Preparations; Pharmacology; Polypharmacy; Predictive Analytics; Proteins; Protocols documentation; Quality of life; ROC Curve; Race; Regimen; Reporting; Research; Research Personnel; Research Proposals; Sampling; Severities; Site; Software Tools; Statistical Data Interpretation; Testing; Tobacco use; Training; United States; Validation; Ventilatory Depression; addiction; adverse drug reaction; career; career development; clinical decision support; clinical practice; clinically significant; cohort; cost; deep learning; deep learning model; design; drug discovery; experience; falls; graph theory; improved; individual patient; knowledge graph; medication-assisted treatment; new therapeutic target; novel; novel therapeutics; opioid epidemic; opioid overdose; opioid use; opioid use disorder; patient safety; pharmacologic; predictive modeling; prescription opioid; relapse patients; side effect; skills; standard of care; sustained recovery; treatment guidelines; vector

PROJECT SUMMARY This proposal for a mentored career development award consists of a training and research plan to facilitate Dr. Zackary Falls' transition to an independent investigator focusing on translational bioinformatics for patient tailored predictive analytics related to opioid addiction severity. The opioid epidemic is a major concern in the United States that is exacerbated due to the high prevalence of prescribing two or more drugs to patients living with opioid use disorder, which increases the likelihood of adverse drug reactions (ADRs) occurring in these patients. Knowing and predicting drug–drug interactions (DDIs) and resulting ADRs is critical for the safety of patients, but ADR prediction software tools used in clinical practice have many limitations. Firstly, most DDI databases used in these software tools are incomplete because they incorporate only pair–wise DDIs. Additionally, most software tools do not incorporate biological mechanism of action information for the drugs and omit relevant patient– specific clinical data such as diagnoses, tobacco use, etc. Dr. Falls aims to exceed the efficacy of these software with the creation of embedded representations for each patient's prescription profile, leveraging both drug–protein interaction knowledge about the prescription drugs and patient level clinical data pertaining to polypharmacy and ADRs. The specific aims of this research are to predict and validate novel off–target proteins for opioids and other commonly co–prescribed medications (Aim 1), extract polypharmacy interactions and ADR relationships from electronic health records of opioid prescription patients (Aim 2), and design a patient personalized software that uses deep–learning architecture to predict severe ADRs caused by opioid related polypharmacy interactions (Aim 3) to be integrated with clinical decision support systems for the benefit of patients and clinicians. The ap- plicant has detailed a rigorous plan containing three career development goals for gaining the skills and expertise to accomplish his research aims. These goals include: Goal 1. Gain knowledge in addiction research and phar- macology as it relates to opioid use, Goal 2. Acquire advanced statistical analysis skills for clinical datasets, and Goal 3. Increase understanding of graph theory and knowledge graph implementation. The team of mentors and collaborators that has been assembled by Dr. Falls, including Prof. Ram Samudrala as primary mentor, perfectly accounts for expertise in research areas that the applicant will be investigating and have knowledge in domains that complement his own understandings to aid in the career development aspect of this proposal. Dr. Falls has the aptitude, creativity, and perseverance to become an excellent researcher. The support of this K01, guidance from his terrific team of mentors and collaborators, and the influence of a rich research environment will enable him to further develop his skills and knowledge. He will surely accomplish all of his career development goals and research aims, become a successful independent investigator, and flourish in his career.

项目总结 这项关于职业发展导师奖的提议包括一项培训和研究计划，以促进Dr。 扎卡里·福尔斯向专注于为患者量身定做的翻译生物信息学的独立研究员的转变 与阿片成瘾严重程度相关的预测性分析。阿片类药物的流行是美国的一个主要问题。 由于给患有艾滋病的患者开两种或两种以上药物的高流行率而加剧的州 阿片类药物使用障碍，这增加了这些患者发生药物不良反应(ADR)的可能性。 了解和预测药物相互作用(DDiS)及其引起的不良反应对患者的安全性至关重要，但 临床上使用的ADR预测软件工具有很多局限性。首先，大多数DDI数据库使用 在这些软件中，工具是不完整的，因为它们只包含成对的DDI。此外，大多数软件 工具没有包含药物的生物作用机制信息，并省略了相关患者- fic临床数据，如诊断、烟草使用等。falls博士的目标是超越这些软件的efficacy。 通过为每个患者的处方PROfiLE创建嵌入的表示，利用药物-蛋白质 关于处方药和患者层面临床数据的互动知识，涉及多药联用和 不良反应。这项研究的目的是预测和验证新的阿片类药物靶外蛋白fic。 其他常用联合处方药物(目标1)，摘录多药联用相互作用和ADR关系 从阿片类药物处方患者的电子健康记录(目标2)，并设计一个患者个性化软件 使用深度学习体系结构来预测阿片类药物相关的多药相互作用引起的严重不良反应 (目标3)与临床决策支持系统集成，以造福患者和临床医生。(3)与临床决策支持系统集成，以造福于患者和临床医生。美联社- Plicant详细制定了一项严格的计划，其中包含三个职业发展目标，以获得技能和专业知识 来完成他的研究目标。这些目标包括：目标1。获得成瘾研究和药物方面的知识。 与阿片类药物使用有关的马克学，目标2。获得临床数据集的高级统计分析技能，以及 目标3：增加对图论和知识图实施的理解。导师团队和 福尔斯博士召集的合作者，包括作为主要导师的拉姆·萨穆德拉拉教授，完美地 说明申请人将研究的研究领域的专业知识，并拥有领域知识 这补充了他自己的理解，以帮助这项提议的职业发展方面。福尔斯博士已经 成为一名优秀研究人员的能力、创造力和毅力。本K01的支持，指导 来自他的Terrific团队的导师和合作者，以及丰富的研究环境的支持将使 让他进一步发展自己的技能和知识。他一定会实现他所有的职业发展目标 和研究的目标，成为一名成功的独立调查员，并在他的事业fl锦上添花。