REU Site: Data Science in the Life Sciences, Environmental Science and Engineering

哈维·穆德学院生命科学、环境科学和工程系数据科学本科生研究体验项目每年将为来自美国各地的9名本科生提供学习数据科学概念和工具并将其应用于(I)生物和生命科学、(Ii)环境科学和(Iii)工程和工业应用项目的机会。学生可以通过对他们最喜欢的项目进行排名来申请REU项目，招聘将考虑申请者的技能和获得成功经验的可能性，以及吸引来自传统上代表性较低的群体的学生，包括女性、非裔美国人和西班牙裔/拉美裔学生。参与的学生将在哈维穆德学院(克莱蒙特的5所学院的成员，还包括克莱蒙特麦肯纳学院、波莫纳学院、皮策学院和斯克里普斯学院)度过总共10周的暑期，并将与一名经验丰富的教职员工一起开展研究项目。它们将得到最先进的基础设施的支持，如计算设施、图书馆和实验室。学生还将与他们的同龄人一起参加一系列数据科学和专业技能研讨会。社交活动和教育实地考察使该项目更加完善。使用数据科学方法和工具的特定领域研究为学生提供了重要的技能，使他们为研究生学习做好准备，并有助于分析和解决许多学科和环境中的问题。专业技能单元，包括研究道德、时间管理和学术出版，将补充和丰富技术培训，并进一步使学生具备成为全面、成功的研究人员所需的能力。随着计算能力的持续快速增长，所有科学家和工程师对数据科学素养的需求都在增加。通过这个REU项目让学生接触到数据科学的相关概念和工具，旨在鼓励他们在STEM相关领域追求职业生涯，并通过培养能够对公共和私营部门的科学和技术产生影响的毕业生，帮助填补美国长期存在的技能和劳动力缺口。这些研究项目是哈维·穆德学院REU生命科学、环境科学和工程学数据科学项目的一部分，利用计算、数学和统计方法和工具解决不同STEM学科中的新问题和公开问题。参与的教师导师在这些轨道上保持积极的研究活动，为探索和深入分析提供丰富的数据和假设环境。实例项目包括：流式细胞仪数据分析；流行病学和公共卫生数据建模，如发展中国家的白内障手术覆盖率；与非常规石油和天然气开发有关的健康风险评估的空间数据建模和分析；测试大气化学中的混合数学模型；以及开发运动教练预测模型，例如根据逐场比赛篮球数据进行实时教练推荐。这些项目涉及高维数据分析，并使用不同的技术来提取洞察力，例如用于验证有效健康干预覆盖的数学模型的模拟数据、多元空间回归和克里格法、关于空气污染模型的云室数据的时间序列分析、部分排序数据的代数分析以及机器学习中用于预测玩家表现的预测技术。此外，学生还将接受R、Python、MatLab的实践培训，精通Linux命令行处理、批处理脚本、数据移动、XSEDE超级计算机的使用和版本控制。他们将接触到Hadoop和Spark等大数据环境，学习使用关系数据库和编写SQL和PostgreSQL查询，并使用ESRI的ArcGIS为高分辨率空间数据建模。虽然技术培训和参与实际研究过程是REU计划的主要组成部分，但也教授成功开发和运行研究计划所需的软技能。这包括参加哈维穆德学院每周成功的Stauffer讲座和开放实验室系列，以及一系列由学院学术部门和写作中心的额外人员参加的定制研讨会。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This Research Experiences for Undergraduates (REU) in Data Science in the Life Sciences, Environmental Science and Engineering at Harvey Mudd College will provide nine undergraduate students per year from across the United States with opportunities to learn and apply data science concepts and tools to projects in the (i) biological and life sciences, (ii) environmental science and (iii) engineering and industrial applications. The students can apply to the REU program by ranking their favorite projects and recruitment will consider the applicant's skills and likelihood of having a successful experience as well as attracting students from traditionally underrepresented population groups, including women, African American and Hispanic/Latino students. The participating students will spend a total of 10 summer weeks at Harvey Mudd College (a member of the 5 Claremont Colleges that also include Claremont McKenna, Pomona, Pitzer and Scripps) and will work on their research projects with an experienced faculty member. They will be supported by state-of-the-art infrastructure such as computing facilities, libraries and laboratories. Students will also engage with their peers in a series of data science and professional skill workshops. Social events and educational field trips round out the program. Domain-specific research using data science methods and tools provide the students with important skills that prepare them for graduate studies and are useful for analyzing and solving problems in many disciplines and environments. The professional skill modules, including research ethics, time management and scholarly publishing, will complement and enrich the technical training and further equip students with competencies needed to become well-rounded, successful researchers. As computational capacity continues to expand at a rapid pace, there is an increased need for data science literacy among all scientists and engineers. Exposing the students to relevant concepts and tools in data science through this REU program is aimed at encouraging them to pursue careers in STEM-related fields and helping fill the persistent skill and labor gap in the U.S. by producing graduates who can have an impact on science and technology in the public and private sectors. The research projects that are part of the Harvey Mudd College REU program in Data Science in Life Sciences, Environmental Science and Engineering address new and open problems in different STEM disciplines using computational, mathematical and statistical methods and tools. The participating faculty mentors maintain active research activities in these tracks that offer data and hypothesis-rich environments for exploration and in-depth analysis. Example projects include the analysis of flow cytometry data, modeling of epidemiological and public health data such as surgical cataract coverage in developing countries, spatial data modeling and analysis for health risk assessments related to unconventional oil and gas development, testing hybrid mathematical models in atmospheric chemistry, and developing predictive models for sports coaching such as real-time coaching recommendations based on play-by-play basketball data. These projects involve high-dimensional data analytics and use different techniques to extract insights such as simulated data to validate mathematical models of effective health intervention coverage, multivariate spatial regression and Kriging, time series analysis of cloud-chamber data on air pollution models, and algebraic analysis of partially ranked data coupled with predictive techniques used in machine learning to predict player performance. In addition, students will also receive hands-on training in R, Python, MATLAB, become well versed in Linux command line processing, batch scripting, data movement, use of XSEDE supercomputers, and version control. They will be exposed to big data environments such as Hadoop and Spark, learn to use relational databases and write SQL and PostgreSQL queries, and work with ESRI's ArcGIS to model high-resolution spatial data. While the technical training and participation in real research processes is the main component of the REU program, necessary soft skills for successfully developing and running research programs are taught as well. This includes the group's participation in Harvey Mudd College's successful weekly Stauffer lecture and open lab series as well as a series of custom-tailored workshops involving additional personnel from the College's academic departments and the Writing Center.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.