The proposed Cold Spring Harbor Laboratory (CSHL) summer course on Statistical Methods for Functional Genomics is to be held annually in 2013-2017. The primary objective of the course is to build competence in statistical methods for analyzing high-throughput data in genomics and molecular biology. In the past decade, high-throughput assays have become pervasive in biological research due to both rapid technological advances and decreases in overall cost. For example, many standard genomic measures such as methylation, copy-number variation, and chromatin immunoprecipitation have been adapted in recent years to high-throughput formats. This has produced an explosion of genome-scale data from multiple organisms, and investigators are now needed who have robust training in relevant statistical methods for analyzing such data. To make meaningful biological inferences from genome-wide data sets, it is essential that both experimental and computational biologists understand the fundamental statistical principles underlying analysis methods. CSHL proposes to meet the need for this specialized, interdisciplinary training by continuing to offer an advanced two-week course each summer entitled Statistical Methods for Functional Genomics. This course will provide intensive, hands-on training that will prepare participants to initiate analyses of large and complex biological data sets. In addition, the curriculum will address issues common to all high-throughput technologies, such as identifying and compensating for systematic errors, statistical significance on a genome-wide scale, and incorporating bioinformatics data into statistical procedures. In-class exercises and demonstrations will be done using the R environment for statistical computing as well as Bioconductor, an open-source project in R for use in bioinformatics research. The course will involve detailed lectures and invited presentations as well as hands-on computer tutorials. The course instructors will be established researchers who are fully active in and have made significant contributions to the analysis of complex biological data sets, and the instructors will be supplemented by a series of invited speakers who will present current research in their fields of expertise to illustrate principles taught in the course. The course will train approximately 24 students per year, ranging from advanced graduate students to senior investigators. Applications are anticipated from scientists with a variety of scientific backgrounds, including molecular evolution, development, neuroscience, cancer, plant biology, and immunology. As with other CSHL postgraduate courses, the overarching goal of Statistical Methods for Functional Genomics is to provide residential training in advanced methodologies that participants can apply immediately to their own research.
Biologists study the behavior and activities of cells at many different scales. For example, a biologist may study how the activities and behaviors are inherited when a cell divides, or when they have been disrupted through mutation or environmental insults in cancer tissues. Technological advances in the past two decades have produced a variety of high-throughput methods for generating vast amounts of data about cells, data that must be analyzed statistically due to the sheer volume and complexity of the measurements involved. The primary objective of the course Statistical Methods for Functional Genomics is to train both experimental and computational scientists in statistical methodology for analyzing high-throughput data sets, so that meaningful biological inferences can be made from them.