BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Sparse discriminative latent characteristics for predicting cancer
  drug sensitivity - David Knowles (Stanford University)
DTSTART:20140110T110000Z
DTEND:20140110T120000Z
UID:TALK49947@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:Various recent experimental studies have involved assaying the
 \nsensitivity of a range of cancer cell lines to an array of anti-cancer\n
 therapeutics. Alongside these sensitivity measurements high\ndimensional m
 olecular characterisation of the cell lines is typically\navailable\, incl
 uding gene expression measurements\, copy number\nvariation and genetic mu
 tations. We propose a sparse multitask\nregression model which learns disc
 riminative latent characteristics\nwhich are highly predictive of drug sen
 sitivity and predictable from\nmolecular features. We use ideas from Bayes
 ian nonparametrics to\nautomatically infer the appropriate number of these
  latent\ncharacteristics. An extension using a binary MRF allows additiona
 l\nprior knowledge\, for example which drugs have shared inhibition\ntarge
 ts\, to be incorporated.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR