Mathematical Sciences/GIG: Graduate & Postdoctoral Education in Cross-Disciplinary Research

通过研究生和博士后教育，统计部力求增加能够在统计与自然科学和社会科学的具体领域之间开展工作的研究工作者的人数。应用的目标领域是遗传学和系谱分析、医学图像分析、大气物理和地球物理科学、天体物理学和金融。在过去的几年里，大量人类谱系的基因数据以爆炸性的速度增长，而且几乎可以肯定，在接下来的十年里，这种速度将会加快。使用多标记对复杂遗传疾病进行连锁分析所涉及的推理和计算问题是非标准的，因此对具有定量统计遗传学知识的统计学家的要求非常高。活跃在芝加哥的团体包括糖尿病方面的格雷姆·贝尔和哮喘方面的卡罗尔·奥贝尔，这两人都与统计部门的奥古斯丁·孔和玛丽·萨拉·麦克佩克有着长期的合作关系。雅丽·阿米特与放射学的陈金图合作，处理各种类型医学图像的模型配准和图像匹配。所使用的技术包括可变形模板和可分解图上的动态编程。形状和图像识别是未来工作的一个挑战。该计划要求对该项目支持的博士后和研究生进行统计遗传学、图像分析、大气或天文现象的时空建模或与应用领域有关的其他专题的培训。就地球物理科学而言，通过采用源自时间序列和空间过程的统计方法来解决平流层臭氧消耗等问题。参与该项目的人员包括来自地球物理科学学院的John Frederick，来自商学院研究生院的George Tiao，来自Argonne的Barry Lesht，以及来自统计学的Michael Stein和肖李蒙。该部的目标是增加受过统计学培训、能够在自然科学和社会科学的若干领域开展合作研究的研究工作者的人数。该计划将招聘统计学或相关学科的博士，并为他们提供必要的培训，以弥合统计学与科学之间的差距。目标应用领域包括数量遗传学、医学图像分析、大气物理和地球物理科学、天体物理和金融。这些领域符合对成功至关重要的三个标准：对更广泛合作的迫切需要，统计系活跃在这些领域的教职员工，以及该大学或阿贡大学的科学家愿意担任导师。过去几年，随着研究人员意识到识别疾病易感基因的潜在好处，广泛人类家系的遗传数据的可用性呈爆炸性增长。随着人类基因组计划的推进，未来十年这一步伐将会加快，因此对具有定量统计遗传学知识的统计学家的需求肯定会增加。就地球物理科学而言，通过采用源自时间序列和空间过程的统计方法来解决平流层臭氧消耗等问题。参与该项目的人员包括来自地球物理科学学院的John Frederick，来自商学院研究生院的George Tiao，来自Argonne的Barry Lesht，以及来自统计学的Michael Stein和肖李蒙。这项活动的资金将由数学科学司和重大计划署多学科活动办公室提供。

Through graduate and postdoctoral education, the Department of Statistics seeks to increase the number of research workers who are capable of working at the interface between statistics and specific areas in the natural and social sciences. The target areas of application are genetics and pedigree analysis, medical image analysis, atmospheric physics and geophysical sciences, astrophysics, and finance. The availability of genetic data on extensive human pedigrees has increased at an explosive rate over the past few years, and the pace is almost certain to increase in the next decade. Inferential and computational problems associated with linkage analysis for complex genetic diseases using multiple markers are non-standard, so the demand for statisticians knowledgeable in quantitative statistical genetics is very high. Groups active at Chicago include Graeme Bell on diabetes, and Carole Ober on asthma, both of whom have long-standing collaborations with Augustine Kong and Mary Sara McPeek from Statistics. Yali Amit's work with Chin-Tu Chen from Radiology deals with model registration and image matching for various types of medical images. The techniques used include deformable templates, and dynamic programming on decomposable graphs. Shape and image recognition is a challenge for future work. The plan calls for postdocs and graduate students supported by this project to be trained in statistical genetics, image analysis, spatio-temporal modelling of atmospheric or astronomical phenomena, or other topics relevant to the area of application. In the case of geophysical sciences, issues such as stratospheric ozone depletion are tackled by adapting statistical methods derived from time-series and spatial processes. The personnel involved in this project include John Frederick from Geophysical Sciences, George Tiao from the Graduate School of Business, Barry Lesht from Argonne, and Michael Stein and Xiao-Li Meng from Statistics. The Department aims to increase the number of research wo rkers trained in statistics who are capable of collaborative research in a number of areas of the natural and social sciences. The plan is to recruit doctorates in statistics or in the relevant discipline and to provide them with the necessary training to bridge the gap between statistics and science. The targeted areas of application include quantitative genetics, medical image analysis, atmospheric physics and geophysical sciences, astrophysics and finance. These are areas that meet three criteria crucial to success: a critical need for more extensive collaboration, faculty in the Department of Statistics active in these areas, and scientists at the University or at Argonne willing to serve as mentors. The availability of genetic data on extensive human pedigrees has increased at an explosive rate over the past few years, as research workers have realized the potential benefits from identifying disease-susceptibility genes. With the progress of the human genome project, the pace will increase in the next decade, so the demand for statisticians knowledgeable in quantitative statistical genetics is certain to increase. In the case of geophysical sciences, issues such as stratospheric ozone depletion are tackled by adapting statistical methods derived from time-series and spatial processes. The personnel involved in this project include John Frederick from Geophysical Sciences, George Tiao from the Graduate School of Business, Barry Lesht from Argonne, and Michael Stein and Xiao-Li Meng from Statistics. Funding for this activity will be provided by the Division of Mathematical Sciences and the MPS Office of Multidisciplinary Activities.