Statistics Colloquium, Dr. Sebastian George

Title: Probability Models for the Analysis of Microarray Gene Expression Data

Abstract: 

Microarrays allow the study of the expression profile of hundreds to thousands of genes simultaneously. These expressions could be from treated samples and the healthy controls. However, microarray expression data are asymmetric, heavy tailed and more peaked compared to the normal and Laplace models. A usual alternative model is the family of asymmetric Laplace models. We propose two families of probability models for modeling microarray expression data. These models are better than normal and Laplace models and the estimation of the parameters can be easily carried out compared to the asymmetric Laplace models. A simulation study is carried out to test the performance of the algorithm.  The maximum likelihood estimation procedure is employed to estimate the parameters of the proposed distributions and an algorithm in R package is developed to carry out the estimation. AIC and BIC criterion are used to compare the distributions.