BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Do modules or pathways help predict breast cancer outcome? - Gunna
 r Klau (Centrum Wiskunde &amp\; Informatica\, Amsterdam)
DTSTART:20110404T150000Z
DTEND:20110404T160000Z
UID:TALK28009@talks.cam.ac.uk
CONTACT:Florian Markowetz
DESCRIPTION:An important topic in systems biology is the identification of
  functional modules in protein-protein interaction networks by means of de
 tecting jointly differentially expressed network regions. The first part o
 f this talk is about an exact integer linear programming solution for subn
 etwork module detection (Dittrich et al.\, 2008).\nRecently\, module and p
 athway based methods have also been proposed for breast cancer outcome pre
 diction (e.g.\, Chuang et al.\, 2007\, Lee et al.\, 2008). The second part
  of this talk addresses the performance of these methods on different esta
 blished data sets using a unified validation procedure. Surprisingly\, and
  contrary to previous claims\, we find that current network or pathway bas
 ed methods do not outperform a simple single gene-based classifier in term
 s of quality and stability.
LOCATION:Cancer Research UK Cambridge Research Institute\, Room 009
END:VEVENT
END:VCALENDAR