BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Automated image analysis for high-throughput cell-based microscopy
  assays with R and Bioconductor -  Dr. Oleg Sklyar\, European Bioinformati
 cs Institute-EMBL
DTSTART:20070314T140000Z
DTEND:20070314T150000Z
UID:TALK6854@talks.cam.ac.uk
CONTACT:Danielle Stretch
DESCRIPTION:Advances in automated microscopy have made it possible to cond
 uct \nlarge-scale cell-based assays with image-type phenotypic readouts. \
 nReliable and reproducible automated image processing and analysis on a \n
 high-throughput scale is one of the main challenges in analysing such \nas
 says. It forms the basis for the consequent statistical analysis and \nass
 essment of biological function. The analysis cycle includes image \nproces
 sing\, image analysis\, phenotype quantification\, statistical \nanalysis 
 of phenotypic data and assessment of biological function with \nthe help o
 f biological databases.\n\nEBImage [1] is a free and open source R-based t
 oolkit for image \nprocessing and analysis designed specifically for autom
 ated image \nanalysis in high-throughput imaging studies. R is a powerful 
 programming \nand scripting language for statistical computing [2]. It is 
 widely used \nfor biological data and databases through the Bioconductor p
 roject [3]. \nCombined with the power of R in machine learning (clustering
  and \nclassification) and hypothesis testing\, and with Bioconductor pack
 ages\, \nEBImage allows to perform the full analysis cycle of imaging data
  from \ncell-based assays. The package supports a wide range of image form
 ats. \nRealized image processing algorithms include image sharpening\, \ns
 egmentation\, edge detection\, morphological operations\, watershed and \n
 distance transfor­mations. Functions for object detection and extractio
 n \nof descriptors like size\, intensity\, a‑circularity etc. are als
 o \navailable along with routines for visualization and quality assessment
 . \nAdditionally\, all mathematical and signal processing algorithms that 
 are \navailable for R can be applied to images.\n\nThe package is used for
  the analysis of genome-wide RNAi microscopy \nscreens. The experiments co
 mprise more than 20000 genes and hundreds of \nthousands of images. We cou
 ld easily identify a subset of genes\, which \nloss of function lead to a 
 particular morphological change\, \nirregularity\, of the cell shape. Gene
 s selected in such a way are now \nbeing investigated further for their ro
 le in a particular cellular pathway.\n\n1. EBImage project page\, www.ebi.
 ac.uk/~osklyar/EBImage/\n2. R Project for Statistical Computing\, www.r-pr
 oject.org\n3. Bioconductor\, www.bioconductor.org\n
LOCATION:MR5\, DAMTP
END:VEVENT
END:VCALENDAR