Interdisciplinary Health Equity Research (IHER) Center

跨学科卫生公平研究(IHER)中心 大连理工大学数据科学研究和教育中心 摘要 加强DS大学的数据科学研究和教育能力，这是一个历史上- 黑人，主要是本科院校将促进多样性，促进创新，促进 合作，并激励子孙后代。这个项目将利用我们RCMI资助的 跨学科卫生公平研究(IHER)中心支持发展更具包容性和公平性的 数据科学社区，利用来自不同背景的个人的才华和观点。我们的 该项目的重点是目标2--开发基础设施以支持数据科学研究、教育和 训练。我们的项目将支持密歇根州立大学新聘用的教员，以及我们的DNA测序核心设施 性能计算(HPC)系统，支持涉及深度学习、神经网络的生物医学研究应用 网络和大数据处理。高性能计算系统中的图形处理单元(GPU)是为并行设计的 并且在执行数学计算时非常高效。他们擅长像矩阵运算这样的任务 以及数据科学研究中常用的深度学习算法。拥有GPU系统可实现更快 计算和训练机器学习模型，减少试验和分析所需的时间。 该系统还将支持大型数据集的研究，例如通过加快数据速度在基因组学中发现的那些数据集 预处理、特征提取和模型训练，使高效处理海量数据集变得更容易。 我们IHER中心的调查员开发核心将举办研讨会和培训，以增强数据- 相关技能和知识，激发兴趣，建立对教师和其他研究人员的信心。培养学生 对于通过PI/学生参与DSU数据科学研究和 教育中心，我们将链接到由蔡斯·斯特拉顿博士(DSU)开发的NSF联合研究培训计划 和盖尔·罗森(德雷克塞尔饰)--《NRT-AI：AI--在特拉华山谷接受理解和操纵的培训》 生物系统(AI-DeV4BIO)。该计划为研究生提供一年的数据科学奖学金， 并将在DSU建立一个数据科学证书，该证书链接到Drexel的现有项目，并包括 共同为两个校区的生物医学研究生开发新课程。证书 该计划将与学生的研究生课程相结合，并为他们提供所需的技能和知识 将数据科学纳入他们的论文或学位论文项目。 我们的项目还将提供基础设施和专业知识来操作、管理和维护我们的高性能计算系统 高性能计算系统有特定的功率和冷却要求，需要高速和可靠的 网络基础设施。此外，管理和维护HPC系统需要一系列专业知识。我们的 项目将确保HPC群集的实验室空间有足够的基础设施，我们将招聘一名HPC 系统管理员处理硬件管理、网络管理、存储管理、故障排除 以及解决问题，以及保持我们的HPC群集运行所需的文档和通信任务。

Interdisciplinary Health Equity Research (IHER) Center DSU Data Science Research and Education Hub Summary Strengthening the data science research and education capacity at DSU, a Historically- Black, primarily undergraduate institution will promote diversity, foster innovation, facilitate collaboration, and inspire future generations. This project will leverage our RCMI-funded Interdisciplinary Health Equity Research (IHER) Center to support the development of a more inclusive and equitable data science community that leverages the talents and perspectives of individuals from diverse backgrounds. Our project is focused on Objective #2 - the development of infrastructure to support data science research, education, and training. Our project will support newly hired faculty at DSU, and our DNA sequencing core facility with a high performance computing (HPC) system to enable biomedical research applications involving deep learning, neural networks and big data processing. GPUs (Graphics Processing Units) in HPC systems are designed for parallel processing and are highly efficient in performing mathematical computations. They excel at tasks like matrix operations and deep learning algorithms, which are commonly used in data science research. Having a GPU system enables faster computation and training of machine learning models, reducing the time required for experimentation and analysis. The system will also support research with large datasets, such as those found in genomics by speeding up data preprocessing, feature extraction, and model training, making it easier to work with massive datasets efficiently. The Investigator Development Core of our IHER Center will sponsor workshops and training to enhance data- relevant skills and knowledge, engage interest and build confidence in faculty and other researchers. To train students for the novel research that will emerge through PI/student engagement in the DSU Data Science Research and Education Hub, we will link to a joint NSF Research Traineeship Program developed by Drs. Chase Stratton (DSU) and Gail Rosen (Drexel) – “NRT-AI: AI-training in the Delaware Valley for Understanding and Manipulating Biological Systems (AI-DeV4BIO).” The program supports one year Data Science Fellowships for graduate students, and will establish a certificate in data science at DSU that links to an existing program at Drexel, and includes the development of new courses jointly available to biomedical graduate students on both campuses. The certificate program will integrate with students’ graduate curriculum and provide them with the skills and knowledge needed to incorporate data science into their thesis or dissertation projects. Our project will also provide the infrastructure and expertise to operate, manage, and maintain our HPC system High-performance computing systems have specific power and cooling requirements and need high-speed and reliable networking infrastructure. In addition, managing and maintaining an HPC system requires a range of expertise. Our project will ensure that the lab space for the HPC cluster has adequate infrastructure, and we will recruit an HPC System administrator to handle hardware management, network administration, storage management, trouble shooting and problem solving, as well as the documentation and communication tasks needed to keep our HPC cluster operating.