[BioC] Limma Script

James W. MacDonald jmacdon at med.umich.edu
Fri May 11 16:04:34 CEST 2007


Shawn Westaway wrote:
> Dear Listers,
> I have a working Limma script below, but I don't know what the output
> is comparing. Can anyone give me a clue? Also, you can see from my run
> that i am having other syntax problems.
> Thanks,
> Shawn
> 
> Script:
> 
> dataMatrix <- exprs(lumi.N)
> presentCount <- pData(featureData(lumi.N))$presentCount
> selDataMatrix <- dataMatrix[presentCount > 0, ]
> selProbe <- rownames(selDataMatrix)
> sampleType <-
> c('1st_litter','WT','2nd_litter','WT','1st_litter','2nd_litter','WT','1st_litter')
> require(limma)
> design <- model.matrix(~ factor(sampleType))
> colnames(design) <- c('1st_litter', '2nd_litter', 'WT')

This is almost certainly wrong. Your model matrix will have an 
intercept, so the first column will be 1st_litter, but the second will 
be 2nd_liter - 1st_litter, and the third will be wt - 1st_litter.

Probably you want to compare the litters to wt, so you might do 
something like

sampleType <- 
factor(c('1st_litter','WT','2nd_litter','WT','1st_litter','2nd_litter','WT','1st_litter'),
	levels=c("WT","1st_litter","2nd_litter"))

Now the second coefficient will be 1st_litter - wt and the third 
coefficient will be 2nd_litter - wt.


> fit <- lmFit(selDataMatrix, design)
> fit <- eBayes(fit)
> if (require(lumiMouseV1) & require(annotate)) {
> geneSymbol <- getSYMBOL(fit$genes$ID, 'lumiMouseV1')
> fit$genes <- data.frame(fit$genes, geneSymbol=geneSymbol)
> }
> topTable(fit, coef='WT', adjust='fdr', number=10)
> p.adj <- p.adjust(fit$p.value[,2])
> sigGene.adj <- selProbe[ p.adj < 0.01]
> sigGene <- selProbe[ fit$p.value[,2] < 0.001]
> }

That is an extra '}', which causes the error below. You also appear to 
be just doing things here, and I don't see what the goal is. Maybe you 
can say what you are trying to do and then we could give some advice.

Best,

Jim


> 
> 
> Transcript:
> 
> 
>>dataMatrix <- exprs(lumi.N)
>>presentCount <- pData(featureData(lumi.N))$presentCount
>>selDataMatrix <- dataMatrix[presentCount > 0, ]
>>selProbe <- rownames(selDataMatrix)
>>sampleType <-
> 
> c('1st_litter','WT','2nd_litter','WT','1st_litter','2nd_litter','WT','1st_litter')
> 
>>require(limma)
> 
> [1] TRUE
> 
>>design <- model.matrix(~ factor(sampleType))
>>colnames(design) <- c('1st_litter', '2nd_litter', 'WT')
>>fit <- lmFit(selDataMatrix, design)
>>fit <- eBayes(fit)
>>if (require(lumiMouseV1) & require(annotate)) {
> 
> + geneSymbol <- getSYMBOL(fit$genes$ID, 'lumiMouseV1')
> + fit$genes <- data.frame(fit$genes, geneSymbol=geneSymbol)
> + }
> 
>>topTable(fit, coef='WT', adjust='fdr', number=10)
> 
>                      ID geneSymbol      logFC         t      P.Value 
> adj.P.Val
> 9576 ooK5UkVej.E737PSw0      Fkbp5  0.6698199  9.658213 6.216342e-06
> 0.06326372
> 5398 faEIr_9OZAUBsX0VLk       Per2  0.6880751  7.431683 4.825718e-05
> 0.24555665
> 9243 oyruyspXqhyohXojdA        Arc -1.0021833 -7.023328 7.395032e-05
> 0.25086413
> 3875 uefVK.bBeKqfv9XzuQ       Xbp1  0.4280023  6.613818 1.155933e-04
> 0.25790396
> 9581 HAyFSJK.n9pHhIQJS4      Dusp1 -1.0741752 -6.530455 1.269007e-04
> 0.25790396
> 6386 lKF2iniB8HnghGhHXk     Slc2a1  0.7937010  6.371030 1.520511e-04
> 0.25790396
> 9717 iUEl382FlILs4VFkFk      Tekt4  0.3838363  6.036845 2.244065e-04
> 0.32625504
> 8764 otfSbQ70_07x405Ptg        Fos -1.0311305 -5.863374 2.762056e-04
> 0.35136802
> 8467 r7fv79XdQPfX3_5FeU       Sox9 -0.6741969 -5.203197 6.317984e-04
> 0.71442364
> 7276 HbaB_NS3aeHQg7ogic     Polr3e  0.3346411  5.079054 7.431893e-04
> 0.75634378
>               B
> 9576 -0.6906876
> 5398 -1.0774567
> 9243 -1.1777686
> 3875 -1.2909444
> 9581 -1.3156866
> 6386 -1.3647262
> 9717 -1.4753395
> 8764 -1.5372394
> 8467 -1.8044508
> 7276 -1.8608493
> 
>>p.adj <- p.adjust(fit$p.value[,2])
>>sigGene.adj <- selProbe[ p.adj < 0.01]
>>sigGene <- selProbe[ fit$p.value[,2] < 0.001]
>>}
> 
> Error: syntax error
> 
> 
> _______________________________________________
> 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.



More information about the Bioconductor mailing list