BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Bayesian view on cryo-EM structure determination - Sjors Scheres
  (LMB)
DTSTART:20111115T160000Z
DTEND:20111115T170000Z
UID:TALK34371@talks.cam.ac.uk
CONTACT:Aleksandra Watson
DESCRIPTION:Three-dimensional structure determination by single-particle a
 nalysis of cryo-electron microscopy (cryo-EM) images requires many paramet
 ers to be determined from extremely noisy data. This makes the method pron
 e to overfitting\, i.e. when structures describe noise rather than signal\
 , in particular near their resolution limit where noise levels are highest
 . To prevent overfitting\, cryo-EM structures are typically filtered using
  ad hoc procedures\, but the tuning of arbitrary parameters may lead to su
 bjectivity in the results. I describe a Bayesian\ninterpretation of cryo-E
 M structure determination\, where smoothness in the reconstructed density 
 is imposed through a Gaussian prior in the Fourier\ndomain. The statistica
 l framework dictates how data and prior knowledge should be combined\, so 
 that the optimal 3D linear filter is obtained without the need for arbitra
 riness and objective resolution estimates may be obtained. Application to 
 experimental data indicates that the statistical approach yields more\nrel
 iable structures than existing methods\, and is capable of detecting small
 er classes in data sets that contain multiple different structures. 
LOCATION:Biochemistry Lecture Theatre\, Sanger Building
END:VEVENT
END:VCALENDAR