 |
BIOL.670.01, BIOL.470.01 - Statistical Analysis for Bioinformatics
|
|
BIOL.670.01, BIOL.470.01 - Statistical Analysis for Bioinformatics
|
COS-BIOL-670 Statistical Analysis for Bioinformatics
This course will investigate some of the statistical methods that have proved useful in analyzing biological information. Examples include Markov models, such as the Jukes-Cantor and Kimura evolutionary models and hidden Markov models, and multivariate models used for discrimination and classification. Class 3, Credit 3 (F)
This is a graduate level biostatistics course open to advanced undergraduates as well as to graduate students. As discussed in the course description, we will the investigate some of the statistical methods that have proved useful in analyzing biological information. These methods form the heart of a field called computational genomics. We will first review statistics, so that you have the tools you need to analyze large volumes of genomic data. For that basic review we will use a text book and you will learn to use either one of the the statistical packages SPSS or PSPP. Some of you may have already learned to do statistics with Minitab. Minitab is not sufficient for the work of this course. In parallel with reviewing statistics you will be assigned important papers on computational genomics to study and report on. Hopefully, we will complete the review and the papers by the middle of the semester, so you can get started on designing and implementing a term project in computational genomics. There will be no midterm exam. You will be evaluated on the basis of assignments and reports at the midterm. At the end of the semester, you will be evaluated on a combination of assignments, reports, your term project and a final examination.
The text for the statistic review is:
The purpose of the text is to ensure that everybody in the course gets up to the same basic level of understanding of statistics. In addition to the text you will also need a computer with a good internet connection and a web cam. On that computer, you should have (or get) a web browser, a spreadsheet program, and a statistical package. You have a choice of two very similar statistical packages: Statistical Package for the Social Sciences (SPSS), or its open software equivalent (PSPP). SPSS costs money (https://www.gnu.org/software/pspp/get.html). The only reason I am aware of to get SPSS instead of PSPP is if you think you may end up looking for a job that explicitly requires experience with SPSS.
To save you expense, we will work from web resources and published papers for much of the course other than the statistics review. However, if you are more comfortable with paper texts and can afford them, the following two books might be helpful to you.
biostat.mc.vanderbilt.edu/wiki/Main/DataSets
http://www.statsci.org/datasets.html
http://www.biostat.jhsph.edu/courses/bio624/datasets/datasets.htm
https://epi.grants.cancer.gov/dac/