Interdisciplinary Research Training in Biostatistics

描述（由申请人提供）：该培训计划旨在培养一批年轻的生物统计科学家骨干，使他们成为理解公共卫生领域生物科学的综合和团队方法的领导者。对许多复杂疾病的易感性可以由修饰基因和环境因素之间的相互作用控制。在后基因组时代，我们开始理解和汇编人类种群中遗传变异的广度。精细地使用这些信息需要发展先进的生物统计分析方法。此外，为了在疾病预防（公共卫生的一个标志）方面取得重大进展，迫切需要将基础科学的进展转化为侧重于预防常见和代价高昂的慢性疾病的规划和政策。只有将新兴的科学信息整合到临床和公共卫生干预措施的设计中，我们才能充分发挥这些进步对公共卫生的价值。已建立的概率和统计推断工具仍然是成功培训新一代生物统计学家的关键，但除此之外，他们必须熟悉在研究中遇到的生物学基础和数据结构，这些研究包括遗传信息、由表达阵列产生的数据，以及寻求在生物系统研究的不同方法中收集的数据之间建立联系。他们还需要在信息科学方面打下坚实的基础，使他们能够利用计算密集型方法，如马尔可夫链蒙特卡罗过程和半参数模型。该培训计划将这些元素与实验室科学和计算生物学的体验式培训相结合，并指导跨学科研究，使毕业生准备好迎接这一挑战。

DESCRIPTION (provided by applicant): This training program is designed to develop a cadre of young biostatistical scientists to become leaders in integrative and team approaches to understanding biological science in the public health arena. Susceptibility to many complex diseases can be governed by the interactions between modifier genes and environmental factors. In the post-genomic era we are beginning to comprehend and compile the breadth of genetic variation within the human population. Refined use of this information requires the development of advanced methods of biostatistical analyses. In addition, to make significant progress in disease prevention, a hallmark of public health, there is a pressing need to translate advances in basic science into programs and policies focused on preventing common and costly chronic diseases. Only by integrating emerging scientific information into the design of clinical and public health interventions can we fully extract the value of these advances for public health. The established tools of probability and statistical inference remain critical to the success in the training of this new generation of biostatisticians, but in addition they must be conversant with the biological basis and data structures encountered in studies that incorporate genetic information, data produced by expression arrays, and that seek to establish connections between data collected in different approaches to the study of biological systems. They will also need a strong grounding in the information sciences preparing them to utilize computationally-intensive methods such as Markov Chain Monte Carlo procedures and semi-parametric models. This training program combines those elements with experiential training in laboratory science and computational biology and directed interdisciplinary research that will prepare graduates to meet this challenge.