BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Robust microarray experiments by design: a multiphase framework - 
 Brien\, C (South Australia)
DTSTART:20110830T153000Z
DTEND:20110830T160000Z
UID:TALK32563@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Speed and Yang with Smyth (2008) outlined six main phases in g
 enomics\, proteomics\nand metabolomics microarray experiments. They sugges
 ted that statisticians could\nassist in the design of experiments in each 
 phase of such an experiment. That being the\ncase\, the experiments potent
 ially involve multiple randomizations (Brien and Bailey\,\n2006) and are m
 ultiphase. Consequently\, a multiphase framework for their design will\nbe
  explored\, the first step of which is to list out the phases in the exper
 iment. One set\nof six phases for the physical conduct of microarray exper
 iments will be described and\nthe sources of variability that affect these
  phases discussed.\nThe multiphase design of an example microarray experim
 ent will be investigated\, beginning with the simplest option of completel
 y randomizing in every phase and then\nexamining the consequences of batch
 ing in one or more phases and of not randomizing\nin all phases. To examin
 e the properties of a design\, a mixed model and ANOVA that\ninclude terms
  and sources for all the identified phases will be derived. For this\, the
 \nfactor-allocation description of Brien\, Harch\, Correll and Bailey (201
 1) will be used.\nIt is argued that the multiphase framework used is flexi
 ble\, promotes the consideration\nof randomization in all phases and facil
 itates the identification of all sources of\nvariability at play in a micr
 oarray experiment.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR