Data Management and Biostatistics

点击翻译按钮获取中文摘要

PROJECT SUMMARY—Data Management and Biostatistics Core (Core C) This Data Management and Biostatistics Core: Core C (Data/Stats Core) is designed to provide high-quality data management infrastructure and statistical support, while using innovation and creativity to improve the delivery of our core functions to investigators. Historically, this PPG has always used sophisticated approaches for phenotyping patients and has prioritized the goal of linking complex clinical syndromes back to neuroanatomic, genetic, and proteomic signatures to improve the early and accurate diagnosis of frontotemporal dementia (FTD). We aim to apply the rich clinical, imaging, molecular, and neuropathological information that we gather on each participant to bridge the knowledge gaps that impede therapeutic discovery in FTD. Thus, the Data/Stats Core for this PPG has a particular mandate to support and facilitate our researchers’ synthesis of complex, multilevel data and to approach that data with appropriately sophisticated recommendations for research design and analytic methods. At the center of the PPG is a comprehensive database built using the LAVA platform, a data management solution specifically designed at UCSF for use by integrative clinical research centers. It provides support for administrative, clinical, and research procedures for central cores and affiliated projects that share common patient cohorts and assessment protocols. In earlier cycles of this PPG, we also developed a suite of browser-based tools designed to work with PPG data and to help solve the problems posed by the complex, multidimensional data generated across PPG projects and cores. These tools facilitate single-patient data integration and normative interpretation across clinical, imaging, specimen, and genetic modalities; visualization of aggregate patient data; rapid linking of phenotypic and genetic data with structural MRI data; and imaging analysis. This core is also structured to provide excellent, comprehensive research design and biostatistical consultation to PPG personnel, guiding researchers with the most appropriate, precise, and innovative statistical approaches. Thus, the aims of this core are: Aim 1: To develop and maintain centralized, integrated data management systems and procedures that ensure the accuracy, availability, and confidentiality of administrative, clinical, and research data from PPG cores and projects. Aim 2: To provide PPG researchers with data integration and analysis tools supporting their ability to effectively combine and utilize data across cores and projects to accelerate discovery. Aim 3: To provide high-quality consultation on biostatistical and data science approaches for all PPG cores and projects in order to systematically unify and focus research design and statistical analysis. Aim 4: To promote research methods integration and collaboration among PPG cores, projects, and related research protocols by promoting Open Science principles and establishing cross-team procedures for efficient sharing of data, models, and code.