BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Non-parametric Bayesian Chromatin State Segmentation  - Will Allen
  (University of Cambridge and MRC-LMB)
DTSTART:20130408T100000Z
DTEND:20130408T110000Z
UID:TALK44041@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:Chromatin state segmentation -- the division of a genome into 
 regions of similar combinatorial patterns of DNA or histone modifications\
 , as measured through high-throughput sequencing -- is a common problem in
  genomics research. Recent large-scale projects have generated enormous am
 ounts of chromatin state information\,\n without corresponding advances in
  techniques for analyzing with these large\, complex datasets.\n\nIn this 
 talk\, I will first outline the problem of chromatin state segmentation an
 d discuss current approaches. I will then review the non-parametric Bayesi
 an "sticky" HDP-HMM model for time-series segmentation\, and introduce a v
 ariational mean-field algorithm for inference in the sticky HDP-HMM with a
 n application to chromatin state segmentation.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR