Hi ? I am very much inexperienced and inexpert in use of BioConductor for analysis of cDNA microarray data. With the help of the listserve (primarily Dr. Gordon Smyth), I have been able to learn to use limma for analysis of data from an equine leukocyte cDNA microarray developed by some friends at the University of Georgia. I have fit a linear model to data from 12 microarray slides, each slide representing hybridization of cDNA from 1 of 12 horses with laminitis against cDNA from a reference sample (pooled cDNA from a group of healthy control horses). When I use topTable to try to identify which genes are differentially expressed, the number with an adjusted P value < 0.05 is quite long (several hundred genes). I?d like to be able to filter the list in various ways; for example, excluding genes whose absolute value of M is < 1. I know that I can do this manually by using the ?sort.by? and ?resort.by? options in topTable, but am looking for an alternative solution that is more automated. Thanks for any help that can be provided. Noah Cohen Noah D. Cohen, VMD, MPH, PhD, DACVIM Professor Department of Large Animal Clinical Sciences College of Veterinary Medicine and Biomedical Sciences Texas A&M University College Station, Texas 77843-4475 Telephone: 979-845-3541 Fax: 979-847-8863 e-mail: ncohen at cvm.tamu.edu

Hi Noah, Noah Cohen wrote: > Hi ? I am very much inexperienced and inexpert in use of BioConductor > for analysis of cDNA microarray data. With the help of the listserve > (primarily Dr. Gordon Smyth), I have been able to learn to use limma > for analysis of data from an equine leukocyte cDNA microarray > developed by some friends at the University of Georgia. I have fit a > linear model to data from 12 microarray slides, each slide > representing hybridization of cDNA from 1 of 12 horses with laminitis > against cDNA from a reference sample (pooled cDNA from a group of > healthy control horses). > > When I use topTable to try to identify which genes are differentially > expressed, the number with an adjusted P value < 0.05 is quite long > (several hundred genes). I?d like to be able to filter the list in > various ways; for example, excluding genes whose absolute value of M > is < 1. I know that I can do this manually by using the ?sort.by? > and ?resort.by? options in topTable, but am looking for an > alternative solution that is more automated. Thanks for any help > that can be provided. I don't know if you would consider this more automated, but something like: ttable <- topTable(fit) ttable <- ttable[abs(ttable[,"M"] > 1,] would do the trick. Best, Jim > > Noah Cohen > > > Noah D. Cohen, VMD, MPH, PhD, DACVIM Professor Department of Large > Animal Clinical Sciences College of Veterinary Medicine and > Biomedical Sciences Texas A&M University College Station, Texas > 77843-4475 Telephone: 979-845-3541 Fax: 979-847-8863 e-mail: > ncohen at cvm.tamu.edu > > _______________________________________________ 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 Noah, My solution is to have a wrapper around topTable where I can give it additional restrictions. Francois topTable2 <- function(fit,contrast=1,p=0.05,p.method='BH', B=NULL,M=NULL,...){ ##can give the contrast as a string if(is.character(contrast)) contrast<-which(colnames(fit$coefficients)==contrast) if(is.null(p)) p <- 1 if(is.null(B)) B <- -Inf if(is.null(M)) M <- 0 answer <- topTable(fit=fit,coef=contrast, number=nrow(fit$coefficients), adjust.method=p.method,...) ##this is necessary because limma changed how topTable work. ##the new versions have a adj.P.Value as well as P.Value if (length(grep('adj.P.Val',names(answer)))==1) pval<-answer$adj.P.Val else pval<-answer$P.Value answer <- answer[(pval <= p & answer$B >= B & abs(answer$M) >= M),] return(answer) } On Wed, 2006-11-29 at 15:59 -0600, Noah Cohen wrote: > Hi ? I am very much inexperienced and inexpert in use of BioConductor > for analysis of cDNA microarray data. With the help of the listserve > (primarily Dr. Gordon Smyth), I have been able to learn to use limma > for analysis of data from an equine leukocyte cDNA microarray > developed by some friends at the University of Georgia. I have fit a > linear model to data from 12 microarray slides, each slide > representing hybridization of cDNA from 1 of 12 horses with laminitis > against cDNA from a reference sample (pooled cDNA from a group of > healthy control horses). > When I use topTable to try to identify which genes are differentially expressed, the number with an adjusted P value < 0.05 is quite long (several hundred genes). I?d like to be able to filter the list in various ways; for example, excluding genes whose absolute value of M is < 1. I know that I can do this manually by using the ?sort.by? and ?resort.by? options in topTable, but am looking for an alternative solution that is more automated. Thanks for any help that can be provided. > Noah Cohen > > Noah D. Cohen, VMD, MPH, PhD, DACVIMProfessorDepartment of Large Animal Clinical SciencesCollege of Veterinary Medicine and Biomedical SciencesTexas A&M UniversityCollege Station, Texas 77843-4475Telephone: 979-845-3541Fax: 979-847-8863e-mail: ncohen at cvm.tamu.edu > _______________________________________________Bioconductor mailing listBioconductor at stat.math.ethz.chhttps://stat.ethz.ch/mailman/list info/bioconductorSearch the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor >