\magnification=1200
\baselineskip=20pt
\nopagenumbers
\font\big=cmr12 scaled \magstep2

\centerline{\bf STANFORD UNIVERSITY}
\centerline{\bf DEPARTMENT OF STATISTICS}
\centerline{\big JOINT STATISTICS/BIOSTATISTICS SEMINAR}
\bigskip

\baselineskip=12pt
\centerline{2:00 p.m., Monday, April 5, 2004}
\centerline{MSOB x303}

\bigskip
\baselineskip=15pt
\centerline{\sl Wing Hung Wong}
\centerline{\sl Department of Biostatistics}
\centerline{\sl Harvard School of Public Health}
\bigskip
\centerline{\bf Some recent progress in microarray analysis}
\bigskip

Abstract:

In this talk I will present two recent results from our group on the 
analysis of microarray data. The first concerns the computation of gene 
expression indexes based on probe-level data from Affymetrix 
oligonucleotide arrays. The method generalizes the probe-level models used 
by software such as dChip and RMA, to explicitly model the effects of 
cross-hybridization of non-target RNAs. We show that by modeling such 
cross-hybridization effects we can improve the accuracy of the expression 
index estimates substantially. The second result is a clustering algorithm 
designed to detect smaller sets of tightly regulated genes from a 
collection of microarray profiles. In contrast to standard clustering 
methods, this "tight cluster" method does not attempt to assign every gene 
to the cluster, and therefore the quality of the resulting clusters will 
not be contaminated by the inclusion of genes with loosely similar patterns.

Information about Workshop in Biostatistics can be found at:
http://www.stanford.edu/dept/HRP/Workshoplist.html





\bye