Model-based Clustering Working Group, Winter 2001: Yeung and Ruzzo Abstract

Working group home page | Adrian Raftery | Statistics Department | University of Washington

Friday, March 2, 2001, from 9:00am to 10:20am in room C-301, Padelford Hall, University of Washington Campus.

Speakers: Ka Yee Yeung and Larry Ruzzo,
Department of Computer Science and Engineering, University of Washington

Title: Model-Based Clustering on Gene Expression Data: Our Current Progress

Abstract:

Before we can apply model-based clustering algorithms to gene expression data, one major question is whether gene expression data satistify the model assumption. In our talk, we are going to report our preliminary experience on testing how well the Gaussian mixture model fits gene expression data, and how much the transformations of taking the logarithm or the square root help to improve normality. In addition, we will also report our experience of using both MCLUST and MBC on transformed gene expression data.