PLACE: 1352 Gilman

SPEAKER: Professor Wei Pan
                    Division of Biostatistics, University of Minnesota
 

TITLE: Statistical Methods for Discovering Differentially Expressed
             Genes in Replicated Microarray Experiments
 

ABSTRACT:

An exciting biological development in the last few years is the use of
microarray technology to simultaneously measure the expression levels of
thousands of genes. Due to its tremendous potential, arguably, microarrays
are called by some researchers as the "array of hope" or "the steam engine
of 21st century".

A remaining challenge is how to extract useful information from the
resulting large amounts of data. An important and common task in analyzing
microarray  data is to identify genes with altered expression under two
experimental conditions. Due to the presence of high noise, it has been widely
recognized that statistical analysis should play an increasingly important
role, such as in determining whether any result obtained is significant or not.
In this talk, I review and compare a classic technique (t-test) and two new
methods, a regression modeling approach (Thomas et al., 2001) and a
mixture model approach (Pan et al, 2001). I will also briefly comment on
the two methods closely related to the mixture model method, the empirical
Bayesian method of Efron et al (2000, 2001) and the Significance Analysis
of Microarray (SAM) method of Tusher et al. 2001).  For illustration, the
methods are applied to the leukemia data of Golub et al (1999).
 

COFFEE: 3:45 p.m., 104 Snedecor Hall