shuffle the sample names of an expressionSet

0

Entering edit mode

Al Tango ▴ 50

@al-tango-3109

Last seen 9.7 years ago

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 colnames(exprs(eset))=sample(colnames(exprs(eset))) rownames(pData(eset))=colnames(exprs(eset))

• 928 views

ADD COMMENT • link updated 15.5 years ago by Martin Morgan 25k • written 15.5 years ago by Al Tango ▴ 50

0

Entering edit mode

Martin Morgan 25k

@martin-morgan-1513

Last seen 27 days ago

United States

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.C EL')) > > # 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

ADD COMMENT • link 15.5 years ago Martin Morgan 25k

0

Entering edit mode

Thanks, Martin, I did find an error in my subsequent process. But your reply is really helpful. I was calculating p-values of analysis using limma. AT On Mon, Nov 10, 2008 at 12:31 PM, Martin Morgan <mtmorgan at="" fhcrc.org=""> wrote: > 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 >

ADD REPLY • link 15.5 years ago Al Tango ▴ 50

Login before adding your answer.