MATH 5070: Topics in Scientific Computation
MATH 5070 - Section 1: Topics in Scientific Computation
Description: The analysis of data sets, large and small, is becoming ubiquitous, as decisions, strategies, and policies will be increasingly data-driven. This is certainly true in biomedicine and health care. As a result, opportunities for professionals with experience in data analytics are plentiful and forecast to increase substantially in the future. Techniques include algorithms from statistics, mathematics, and computer science, ranging from statistical learning techniques to tools from advanced mathematics, such as topology. This course is intended to provide a hands-on introduction to data analytics in the fields of biomedicine and health care, using a "learning by doing" approach. Participants will learn quantitative techniques through immersion in ongoing data analysis projects at both the Storrs and Farmington UConn campuses, as well as the Jackson Laboratory for Genomic Medicine on the Farmington campus. During periodic class meetings, students will present their individual projects, the analysis techniques used, and the challenges encountered. Students may need to spend a minimum of 3-5 hours per week at the Health Center. Course participants should have completed the requirements for the M.S. in Mathematics or Statistics or the equivalent. Consent of the instructors is required. Two examples of projects include: 1. Analysis of ChIP-seq data. Modern high-throughput sequencing technologies have provided inexpensive access to high-quality and high-quantity data of molecular sequences of DNA, RNA, and proteins. These data have made it possible to study events of gene transcription in cells in unprecedented detail, revealing insights into causes of diseases as well as possible ways to treat them. One such technology is ChIP sequencing, which is very useful in analyzing interactions of proteins with DNA. Participants will have an opportunity to analyze data from mouse experiments that focus on the potential of the immune system to help fight cancer. No prior knowledge of molecular biology or bioinformatics is required. 2. Analysis of health insurance claims data. Insurance claims, whether to a private insurance provider or public programs, such as Medicare and Medicaid, can be used to obtain information about a variety of questions, such as efficacy of a particular treatment, prevalence of a particular disease, variation in providers, and serve as the basis of policy decisions and more general health care studies. Participants will be part of a team that analyzes such data for policy support purposes. If you have questions or would like to register, please send a letter of application (maximum two pages) outlining your background and your particular areas of interest to Reinhard Laubenbacher (email@example.com) or Jeremy Teitelbaum (Jeremy.firstname.lastname@example.org) by ?date?
Prerequisites: Consent of instructor.
Sections: Spring 2015 on Storrs Campus
|14664||5070||001||Lecture||W 4:00:00 PM-5:00:00 PM||LH309||Laubenbacher, Reinhard & Teitelbaum|