[BioC] shuffle the sample names of an expressionSet

[Martin Morgan]

Hi Al --

"Al Tango" <time.is.flying at gmail.com> writes:

> Hi, all, I want to randomly shuffle sample names in an object of
> expressionSet, aiming
> to perform permutation test. I did as follows, but the following test
> result remained unchanged (When I shuffled the original data file
> names before 'ReadAffy()' , the result did change). So it must be
> wrong; what's the way please?  Thanks to everybody who help / has
> helped me.
> e.g.
>
> library(affy)
> eset=rma(ReadAffy(filenames=c('sample1.CEL','sample2.CEL','sample3.CEL'))
>
> # below was what I did to shuffle sample names
> 
> 

This

> data(sample.ExpressionSet)
> obj1 <- obj <- sample.ExpressionSet
> sampleNames(obj1) <- sample(sampleNames(obj))

shuffles exprs and phenoData names.

> head(exprs(obj), 2)
                      A        B        C        D       E       F        G
AFFX-MurIL2_at  192.742  85.7533 176.7570 135.5750 64.4939 76.3569 160.5050
AFFX-MurIL10_at  97.137 126.1960  77.9216  93.3713 24.3986 85.5088  98.9086
                      H       I        J       K       L       M       N
AFFX-MurIL2_at  65.9631 56.9039 135.6080 63.4432 78.2126 83.0943 89.3372
AFFX-MurIL10_at 81.6932 97.8015  90.4838 70.5733 94.5418 75.3455 68.5827
                      O       P        Q       R       S       T         U
AFFX-MurIL2_at  91.0615 95.9377 179.8450 152.467 180.834 85.4146 157.98900
AFFX-MurIL10_at 87.4050 84.4581  87.6806 108.032 134.263 91.4031  -8.68811
                       V       W        X       Y        Z
AFFX-MurIL2_at  146.8000 93.8829 103.8550 64.4340 175.6150
AFFX-MurIL10_at  85.0212 79.2998  71.6552 64.2369  78.7068
> head(exprs(obj1), 2)
                      S        G        W        L       C       B        I
AFFX-MurIL2_at  192.742  85.7533 176.7570 135.5750 64.4939 76.3569 160.5050
AFFX-MurIL10_at  97.137 126.1960  77.9216  93.3713 24.3986 85.5088  98.9086
                      N       Q        O       K       R       U       J
AFFX-MurIL2_at  65.9631 56.9039 135.6080 63.4432 78.2126 83.0943 89.3372
AFFX-MurIL10_at 81.6932 97.8015  90.4838 70.5733 94.5418 75.3455 68.5827
                      M       Z        E       F       Y       D         V
AFFX-MurIL2_at  91.0615 95.9377 179.8450 152.467 180.834 85.4146 157.98900
AFFX-MurIL10_at 87.4050 84.4581  87.6806 108.032 134.263 91.4031  -8.68811
                       T       P        X       A        H
AFFX-MurIL2_at  146.8000 93.8829 103.8550 64.4340 175.6150
AFFX-MurIL10_at  85.0212 79.2998  71.6552 64.2369  78.7068
> head(pData(obj), 2)
     sex    type score
A Female Control  0.75
B   Male    Case  0.40
> head(pData(obj1), 2)
     sex    type score
S Female Control  0.75
G   Male    Case  0.40

but I think it is just a tidier way of doing what you already
did. Perhaps what you meant to do was to shuffle some aspects of
phenoData, e.g.,

> obj1 <- obj
> pData(obj1)[["type"]] <- sample(pData(obj1)[["type"]])
> head(pData(obj), 5)
     sex    type score
A Female Control  0.75
B   Male    Case  0.40
C   Male Control  0.73
D   Male    Case  0.42
E Female    Case  0.93
> head(pData(obj1), 5)
     sex    type score
A Female    Case  0.75
B   Male    Case  0.40
C   Male    Case  0.73
D   Male Control  0.42
E Female    Case  0.93

? What are you doing that you would like to see the permutation of?

Martin

> _______________________________________________
> 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

-- 
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M2 B169
Phone: (206) 667-2793