Bayesian Logic Regression for SNP Data

Advances in genetic sequencing technology make it possible to study the effect of genetic variations on the risk of developing a complex disease, such as cancer. The most common form of genetic variations are the Single Nucleotide Polymorphisms (SNPs). Logic regression is a powerful statistical tool for identifying interactions of SNPs that might be associated with a disease. Logic regression however only finds a single best fitting model for the data, while usually there are many models that fit almost equally well. The author Arno Fritsch shows how by embedding logic regression in the framework of Bayesian statistics this drawback can be overcome. The book gives an introduction to both logic regression and Bayesian statistics and shows how the two can be combined. It is demonstrated that considering more than one plausible logic regression model can substantially improve predictions of the disease status. The book is intended for bioinformaticians, epidemiologists and statisticians involved in analyzing SNP data.