An embedded and charset-unspecified text was scrubbed... Name: not available Url: https://stat.ethz.ch/pipermail/bioconductor/attachments/20060420/ c95e79fd/attachment.pl

Density plots look like they fall into 2-3 groups; with a main peak about 6.5 (n=~12), and then 4-5 samples at 7. I'm running the GCRMA/ PCA at the moment. Thanks Tarun -----Original Message----- From: James W. MacDonald [mailto:jmacdon@med.umich.edu] Sent: Thursday, April 20, 2006 1:10 PM To: Tarun Nayar Cc: bioconductor at stat.math.ethz.ch Subject: Re: [BioC] normalizing with different RNAdeg slopes Hi Tarun, Tarun Nayar wrote: > Hi all, > > I'm dealing with a number of human lymphoma samples, some of which > have been in freezer storage for up to 20 years. The samples have been > run on HGU133Aplus2 chips, and have a wide range of 5' to 3' > slopes (as determined by the RNAdeg function), ranging from 3.8 to > 9.5. > > My understanding is that normalizing chips with such a wide range of > slopes could lead to innacuracies downstream. > > 1) Are these slopes too different to allow for the chips to be > normalized together? It's possible, but I find that the density plots tend to dictate better how well things will normalize. What do those look like? > > 2) if not, how would one go about working around these slope > differences? Our PI is dead set on extracting any info possible from > these 'rare' samples. I would run RMA or GCRMA on the data, and then do a PCA plot to see how replicates are grouping. Clusters of replicated samples usually indicate that samples are (in general) fairly similar, whereas widely dispersed replicate samples usually indicate that there is a lot of variability that may not be biological in nature. Best, Jim > > many thanks > > Tarun Nayar Genome Sciences Centre Vancouver, BC > > [[alternative HTML version deleted]] > > _______________________________________________ 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 -- James W. MacDonald, M.S. Biostatistician Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 ********************************************************** Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.

Hi Tarun, Tarun Nayar wrote: > Hi all, > > I'm dealing with a number of human lymphoma samples, some of which > have been in freezer storage for up to 20 years. The samples have > been run on HGU133Aplus2 chips, and have a wide range of 5' to 3' > slopes (as determined by the RNAdeg function), ranging from 3.8 to > 9.5. > > My understanding is that normalizing chips with such a wide range of > slopes could lead to innacuracies downstream. > > 1) Are these slopes too different to allow for the chips to be > normalized together? It's possible, but I find that the density plots tend to dictate better how well things will normalize. What do those look like? > > 2) if not, how would one go about working around these slope > differences? Our PI is dead set on extracting any info possible from > these 'rare' samples. I would run RMA or GCRMA on the data, and then do a PCA plot to see how replicates are grouping. Clusters of replicated samples usually indicate that samples are (in general) fairly similar, whereas widely dispersed replicate samples usually indicate that there is a lot of variability that may not be biological in nature. Best, Jim > > many thanks > > Tarun Nayar Genome Sciences Centre Vancouver, BC > > [[alternative HTML version deleted]] > > _______________________________________________ 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 -- James W. MacDonald, M.S. Biostatistician Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 ********************************************************** Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.