Hi, I am trying to understand how exactly the gcrma software works, and working my way through the Wu/Irizarry paper "A Model Based Background Adjustment for Oligonucleotide Expression Arrays" . My problem is in understanding the Bayes approach that is behind it. I am particularly interested in how the integrated functions look like. Could anyone point me towards a source of the functions or a source with more detail than the paper I mentioned above? Any help would be very much appreciated, Best regards, Anne Pohrt

Anne, People were kind enough to help me with basically the same question, so that I will hazard a summary and let more knowledgeable people add or correct if necessary, In RMA the of the frequency of the intensity of individual probes are plotted against intensity. It is assumed that the noise varies as a Gaussian function and that the signal varies as an exponential. The plot is fit to a Gaussian which is subtracted out, leaving only the signal. In GCRMA this is done separately for each set of probes with a given GC content. Then the distribution of probes on different chips are normalized to give (almost) the same distribution using quantile normalization. Finally the intensity of a probeset is obtained from its component probes by fitting to a linear model using a robust method called medium polish. I hope this makes sense. There are people out there who can correct me if necessary. Best wishes, Rich On Aug 24, 2006, at 3:19 AM, Anne Pohrt wrote: > Hi, > > I am trying to understand how exactly the gcrma software works, and > working my way through the Wu/Irizarry paper "A Model Based Background > Adjustment for Oligonucleotide Expression Arrays" . > My problem is in understanding the Bayes approach that is behind it. I > am particularly interested in how the integrated functions look like. > Could anyone point me towards a source of the functions or a source > with more detail than the paper I mentioned above? > > Any help would be very much appreciated, > Best regards, > Anne Pohrt > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > ------------------------------------------------------------ Richard A. Friedman, PhD Associate Research Scientist Herbert Irving Comprehensive Cancer Center Oncoinformatics Core Lecturer Department of Biomedical Informatics Box 95, Room 130BB or P&S 1-420C Columbia University Medical Center 630 W. 168th St. New York, NY 10032 (212)305-6901 (5-6901) (voice) friedman at cancercenter.columbia.edu http://cancercenter.columbia.edu/~friedman/ "My novel has a hundred characters and it takes place over a year. When I finish it, if it is shorter than Ulysses, then Joyce described things too much". -Rose Friedman, age 9