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

\centerline{\bf STANFORD UNIVERSITY}
\centerline{\bf DEPARTMENT OF STATISTICS}
\centerline{\big DEPARTMENTAL SEMINAR}
\bigskip

\baselineskip=12pt
\centerline{4:15 p.m., Tuesday, January 14, 2003}
\centerline{Sequoia Hall Room 200}
\centerline{(Cookies at 3:45 in 1st Floor Lounge)}

\bigskip
\baselineskip=15pt
\centerline{\sl Jun S. Liu}
\centerline{\sl Department of Statistics}
\centerline{\sl Harvard University}
\bigskip
\centerline{\bf Towards the Integration of Sequence Motif
Discovery and Microarray Analysis}
\bigskip

Although numerous approaches for using microarray data
to help finding  DNA regulatory binding motifs have been
proposed in the past years, available methods either require 
examining too many motifs, which results in possibly many false 
positive results, or use motif-finding  methods without adequate 
specificity to determine significant  motifs. We developed a new
method Motif Regressor for alleviating some of the shortcomings 
of existing methods.  Our method makes use of a newly developed 
motif finding approach MDscan to provide a set  of candidate 
regulatory motifs in the upstreams of 
genes that are most overexpressed (or underexpressed) in an 
experimental condition (or cluster). These putative motifs are
re-examined with a regression analysis to further utilize the 
microarray information.  I will illustrate the application of this
method on several yeast microarray experiments.  I will also 
discuss some issues related to the motif analysis such as 
motif clustering, database motif search, etc.

\bye