[BioC] Limma: topTable without eBayes?

[Jenny Drnevich]

Hi Guido,

topTable for a single coef needs the $t, $p.value and $lods that are 
created from eBayes. You could manually compute these to be regular 
t-statistics instead of those from eBayes. See section 10 of the 
limmaUsersGuide(). Try this (I didn't test it):

fit2$t <- fit2$coef/fit2$stdev.unscaled/fit2$sigma

fit2$p.value <- 2 * pt(-abs(fit2$t), df = fit2$df.residual)

The $lods is harder; if you care about it you'll have to figure out 
how to duplicate it from the ebayes code by using modified inputs, 
which is what I did for the $t and $p-value; if you don't care about 
it but need something for topTable to work, I'd just do:

fit2$lods <- fit3$lods

HTH,
Jenny


At 12:51 PM 8/31/2010, Hooiveld, Guido wrote:
>Dear list,
>
>topTable is a convenient function for displaying the top regulated 
>genes in an experiment. topTable is usually applied on a object that 
>is generated after fitting a linear model AND subsequent application 
>of the eBayes funtion (fit3 below).
>I would like to view the results of a model WITHOUT applying the 
>eBayes function, preferably using topTable. That is, using topTable 
>to summerize the results of fit2 below. Applying topTable as such 
>won't work; apparently a format imposed by the eBayes function is 
>required to have topTable properly work. Even specifying a contrast 
>of interest doesn't work.
>Therefore, has anyone a suggestion how to summerize the results of 
>the fit2 mentioned below?
>
>Thanks,
>Guido
>
> > library(affy)
> > library(limma)
> > targets <- readTargets("targets_A23.txt")
> >
> > affy.data <- ReadAffy(filenames=targets$FileName)
> > x.norm <- rma(affy.data)
>Background correcting
>Normalizing
>Calculating Expression
> >
> > TS <- paste(targets$Strain, targets$Treatment, sep=".")
> > TS <- factor(TS, levels=c("WT.Con","WT.WY","KO.Con","KO.WY"))
> > design <- model.matrix(~0+TS)
> > colnames(design) <- levels(TS)
> > fit <- lmFit(x.norm, design)
> > cont.matrix <- makeContrasts(A=WT.WY-WT.Con, B=KO.WY-KO.Con, levels=design)
> >
> > fit2 <- contrasts.fit(fit, cont.matrix)  #<<---- how to 
> conviently summerize results of this fit?
> >
> > fit3 <- eBayes(fit2)
> >
> > topTable(fit3)
>                 ID        A            B  AveExpr         F 
> P.Value    adj.P.Val
>13289 1431833_a_at 3.538370 -0.091190845 7.856618 1453.7786 
>5.084771e-18 1.153734e-13
>18706 1450643_s_at 2.847376  0.182291682 7.907519 1234.5101 
>1.739465e-17 1.620566e-13
>17177   1449065_at 4.528581  0.125074564 6.713129 1200.7506 
>2.142661e-17 1.620566e-13
>7120  1422925_s_at 3.149781 -0.007867198 8.072523  895.1004 
>1.945926e-16 1.103827e-12
>7192  1422997_s_at 4.298410  0.090012125 8.311977  837.9040 
>3.193016e-16 1.448991e-12
>13047 1431012_a_at 2.682180  0.228653000 8.911287  785.9448 
>5.159295e-16 1.698832e-12
>6721    1422526_at 2.772829  0.123117132 8.060643  784.2985 
>5.240997e-16 1.698832e-12
>16494   1448382_at 3.386068  0.193350865 9.658716  685.0981 
>1.442685e-15 3.703026e-12
>296     1415965_at 3.749143  0.071163645 5.293939  683.4572 
>1.468807e-15 3.703026e-12
>8911    1424716_at 2.647513  0.370083995 8.453969  665.2223 
>1.798165e-15 4.080035e-12
> > topTable(fit2)
>Error in topTableF(fit, number = number, genelist = genelist, 
>adjust.method = adjust.method,  :
>   F-statistics not available. Try topTable for individual coef instead.
> > topTable(fit2, coef=1)
>Error in dim(data) <- dim : attempt to set an attribute on NULL
> >
> > sessionInfo()
>R version 2.11.0 (2010-04-22)
>i386-pc-mingw32
>
>locale:
>[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United 
>States.1252    LC_MONETARY=English_United States.1252 
>LC_NUMERIC=C                           LC_TIME=English_United States.1252
>
>attached base packages:
>[1] stats     graphics  grDevices utils     datasets  methods   base
>
>other attached packages:
>[1] moe430acdf_2.6.0 limma_3.4.0      affy_1.26.1      Biobase_2.8.0
>
>loaded via a namespace (and not attached):
>[1] affyio_1.16.0         preprocessCore_1.10.0 tools_2.11.0
> >
>
>
>------------------------------------------------
>Guido Hooiveld, PhD
>Nutrition, Metabolism & Genomics Group
>Division of Human Nutrition
>Wageningen University
>Biotechnion, Bomenweg 2
>NL-6703 HD Wageningen
>the Netherlands
>tel: (+)31 317 485788
>fax: (+)31 317 483342
>internet:   http://nutrigene.4t.com<http://nutrigene.4t.com/>
>email:      guido.hooiveld at wur.nl
>
>
>
>         [[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