[BioC] questions about altcdfenv

Marc Carlson mcarlson at fhcrc.org
Thu Oct 2 18:39:20 CEST 2008


Hi Hui-Yi,

Without specific examples I can only guess, but I am betting that the
trouble you are having is related to two facts:

1) approximately half of the probesets on the yeast2 affymetrix platform
are from S.pombe. (the other half are from S.cerevisiae)
and
2) the annotation packages that we have made for yeast platforms are
based on SGD, which is important since SGD does not have annotations for
S.pombe.

If you know about some great sources of annotations for S.pombe, I would
love to know more about them.


  Marc



Hui-Yi Chu wrote:
> Dear BioC list,
>
> Before turning to use altcdfenv package, I used genefilter package to
> generate a "filtered" expressionSet, but after filtration and annotation, it
> still contained lots of blank gene symbols. Some of those probesets without
> symbols are from S.Pombe, so that is the reason going back to use
> altcdfenvs. Furthermore, I have checked the annotated+expression value table
> (before filtration, total=10928 probesets), it has abnormally high numbers
> of NA gene names--- 4707...I checked ORF as well, which gave me 5222 blank
> ORFs....
> Please, can anybody give me any idea that how can I deal with this issue??
> Many many thanks!!!
>
> Best,
> Hui-Yi
>
>
> *Codes*:
> ## data importing
>
>   
>> library("affy")
>> library("limma")
>> targets <- readTargets("fed_total.txt")
>> aaa <- ReadAffy(filenames = targets $ filename)
>> eset <- rma(aaa)
>>
>>     
>
> #### add annotation info and output as a table
>
>   
>> library("annotata")
>> library("yeast2.db")
>> annotation(eset) <- "yeast2.db"
>>
>>     
>               #skip following codes for creating a data frame for exporting
> to excel...
>
> ### filtration
> library("genefilter")
> f1 <- anyNA
> f2 <- pOverA(0.25, log2(100))
> ff <- filterfun(f1, f2)
> selected <- genefilter(eset, ff)
> esetsub <- eset[selected,]
>
> ## limma fit and contrast
> library("limma")
> design <- model.matrix(~0+factor(c(1,1,2,2,3,3)))
> colnames(design) <- c("wt", "mut.a","mut.b")
> fit <- lmFit(exprs(esetsub), design)
> fit2 <- eBayes(fit)
> contrast.matrix <- makeContrasts("mut.a vs wt" = mut.a-wt,
>                                 "mut.b vs wt" = mut.b-wt,levels=design)
>     #skip following codes for DEG analysis
>
> 	[[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
>
>



More information about the Bioconductor mailing list