Title: Subspace Clustering for Microarray Data Analysis: Multiple Criteria and Significance Assessment (joint work with Lei Liu and ChengXiang Zhai)

Slide Set [PDF]

Abstract: The goal of microarray data analysis is to identify a subset of  genes whose expression levels rise and fall coherently under a subset of  conditions, that is, they exhibit fluctuation of a similar shape when  conditions change. Typically, clustering algorithms are first used to  group together genes with similar patterns of expression. After  clustering, people would usually inspect the clusters manually with  annotation of genes and assign functions to clusters. Two major challenges  are thus (1) to find gene clusters that are truly meaningful  biologically; and (2) to rank and prioritize the clusters in some  reasonable order for human inspection. We present a new clustering method that addresses both  challenges.

Bio: Hui Fang is a fifth-year Ph.D. Student in the Department of Computer  Science at the University of Illinois. Her research interests include  information retrieval and bioinformatics.