BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Alignment of Unlabelled Marked Point Sets Using Random Fi
 elds - Ian Dryden\, University of Nottingham
DTSTART:20101117T141500Z
DTEND:20101117T151500Z
UID:TALK26667@talks.cam.ac.uk
CONTACT:Rachel Fogg
DESCRIPTION:In structural bioinformatics and chemoinformatics it is of gre
 at interest to align molecules\, but the task is often very difficult. Sta
 tistical methodology is proposed for comparing unlabelled marked point set
 s\, with an application to aligning steroid molecules in chemoinformatics.
  Methods from statistical shape analysis are combined with techniques for 
 predicting\nrandom fields in spatial statistics in order to define a suita
 ble measure of similarity between two molecules. Bayesian modelling of the
  predicted \nfield overlap between pairs of molecules is proposed\, and po
 sterior inference of the alignment is carried out using Markov chain Monte
  Carlo simulation. By representing the fields in reproducing kernel Hilber
 t spaces\, the degree of molecule overlap can be computed without expensiv
 e numerical integration. Superimposing entire fields rather than the confi
 guration matrices of point co--ordinates thereby avoids the problem that t
 here is usually no clear one--to--one correspondence between the atoms. Us
 ing a similar concept\, we also propose an adaptation \nof the generalized
  Procrustes analysis algorithm for the simultaneous alignment of multiple 
 point sets. The methodology is illustrated with a simulation study and the
 n applied to the dataset of 31 steroid molecules\, where the relationship 
 between shape and binding activity to the corticosteroid binding globulin 
 receptor is explored.\n \nThis is joint work with Irina Czogiel and Chris 
 Brignell.\n\n \n
LOCATION:LR4\, Engineering\, Department of
END:VEVENT
END:VCALENDAR