Statistics 697BD - ST-Biomed&HealthDataAnalysis

In this course, we will apply several novel machine learning algorithms, including normalization methods, classification and regression analysis on cancer patient data sets to arrive at personalized cancer treatments. We will develop several supervised learning algorithms for analyzing gene expression data sets. We will review, develop, and evaluate deconvolution methods as applied to gene expression data of primary tumors to predict the percentage of various cell types in tumors. We will implement most of these methods in Python. Although programming skills, machine learning, or computational biology background are preferred, they are not required for this course.

Knowledge of calculus and linear algebra are required.
Knowledge of statistics (Stat 516, Stat 608, or equivalent) preferred, but not required.
Knowledge of regression (Stat 525, Stat 625, Stat 697R, or equivalent) preferred, but not required.

Advanced undergraduate students may request permission of instructor to enroll.