Core E: Data Sciences Core

IDD中几乎每个项目的成功和影响都取决于统计技术的正确使用。因此， 核心E在促进国际发展研究中心所有调查人员的研究以及 其他IDDRC核心和签名研究项目。核心E执行IDDRC的独特功能 调查人员，因为它帮助他们确定和使用统计和方法的专门知识和资源 适用于范德比尔特大学(VU)和范德比尔特大学医学中心(VUMC) 对于他们的问题-特别是对于更复杂的研究设计(例如，多层嵌套)或 具有统计限制(例如，稀有人群研究中常见的小样本量)。此外，通过 临床翻译和翻译神经科学核心B和C的生成性活动，核心E提供 针对IDD相关科学问题量身定做的复杂和非平凡的统计方法和模型(例如， 用于神经成像分析的贝叶斯时空模型)。除了在以下方面拥有丰富的专业知识外 生物统计学、神经统计学和量化心理学，范德比尔特也是发展大型 数据结构和挖掘这些数据以推进健康和发展研究，包括合成 衍生品(SD)，这是从总计超过280万人中收集的电子健康记录数据的未识别数据集 唱片。尽管这样的大数据结构对Vanderbilt，特别是IDDRC来说是令人难以置信的资源 研究人员有能力捕获大量罕见疾病样本，因此将数据放入 可分析的格式，并选择适当的统计方法进行分析。核心E支持IDDRC调查人员 通过三个目标充分利用所有这些VU/VUMC资源：Aim 1，它提供访问 采用现代统计和数据科学方法回答与缺铁性贫血相关的问题，包括开展 对标志性IDDRC研究项目的数据分析；目标2，加强对以下方面的IDD研究的培训 那些从事数据科学方法的人，包括实施一项新的内部培训资助计划 数据科学研究所学员和国际数据发展中心之间的合作；以及支持卫生领域创新的AIM 3- 通过促进使用大型数据集(如SD)进行相关IDD研究，包括提供尖端技术 IDDRC调查人员可以使用的大规模SD IDD管理的数据库的咨询和工具 用于生成试点数据和进行研究。总体而言，Core E的目标和创造性工作以及 与IDDRC其他核心的交互不仅满足了IDDRC调查人员的直接需求，而且还 通过允许开发新的资源、平台和方法来预测未来的情况。通过攻克 并解决复杂的、多模式的数据科学问题，Core E准备做出重大贡献 在接下来的5年里，加快科学发现，以改善IDDS患者的结局。

The success and impact of nearly every project in IDD hinges on the proper use of statistical techniques. Thus, Core E has a critical role in facilitating research for all IDDRC investigators, as well as for the progress of the other IDDRC Cores and Signature Research Project. Core E performs a unique function for IDDRC investigators as it helps them identify and use the statistical and methodological expertise and resources available at Vanderbilt University (VU) and Vanderbilt University Medical Center (VUMC) that are appropriate for their questions – especially for more complicated research designs (e.g., many layers of nesting) or those with statistical limitations (e.g., small sample sizes common in research with rare populations). Further, through generative activity with Clinical Translational and Translational Neuroscience Cores B and C, Core E provides sophisticated and non-trivial statistical methods and models tailored to IDD-related scientific questions (e.g., Bayesian spatio-temporal models for neuroimaging analysis). In addition to having considerable expertise in biostatistics, neuro-statistics, and quantitative psychology, Vanderbilt is also a national leader in developing big data structures and mining that data to advance health and development research, including the Synthetic Derivative (SD), a de-identified dataset of electronic health record data collected from over ~2.8 million total records. Though such big data structures are incredible resources to Vanderbilt, and especially IDDRC investigators with their ability to capture large samples of rare disorders, it can be challenging to put the data in analyzable formats and select suitable statistical approaches for analysis. Core E enables IDDRC investigators to fully capitalize on all these VU/VUMC resources through three aims: Aim 1, which provides access to modern statistical and data science methods to answer questions of relevance to IDD, including conducting data analyses for the Signature IDDRC Research Project; Aim 2, which enhances training in IDD research for those engaging in data science methods, including implementing a novel internal training grant program between Data Sciences Institute trainees and the IDDRC; and Aim 3, which supports innovation in health- related IDD research by facilitating use of large data sets such as the SD, including providing cutting-edge consultations and tools for working with large-scale SD IDD-curated database that IDDRC investigators can use for generating pilot data and conducting studies. Collectively, Core E’s aims and generative work and interactions with other IDDRC Cores not only meets the immediate needs of IDDRC investigators, but also anticipates future ones, by allowing for novel resources, platforms, and methods to be developed. By tackling and solving complex, multi-modal data science questions, Core E is poised to contribute substantially over the next 5 years to accelerating scientific discovery to improve the outcomes of people with IDDs.