[R] Why SAM has totally diffent results in R2.1.1 and R2.4.0
1
0
Entering edit mode
@martin-morgan-1513
Last seen 14 days ago
United States
Wei Zhao This is not my software package and I do not use it regularly, but the following might help in getting help from others: * Direct any follow-up to the bioconductor mailing list * Make sure that you are using a coherent set of current software. * Be more precise than 'totally different'. * Clearly indicate the software you are using. Following my own advice, I tried > update.packages() to update the installed packages; I might have done > source("http://bioconductor.org/biocLite.R") > biocLite() on a new installation. I then ran your suggested code (with library(siggenes) rather than multtest), with the following output: > library(siggenes) Loading required package: multtest Loading required package: Biobase Loading required package: tools Welcome to Bioconductor Vignettes contain introductory material. To view, type 'openVignette()' or start with 'help(Biobase)'. For details on reading vignettes, see the openVignette help page. Loading required package: survival Loading required package: splines Changes in siggenes: Version 1.2.x: Added sam2excel, sam2html, and help.sam. Version 1.5.3: Now exprSet objects can also be used in the EBAM functions for specifying 'data' and 'cl'. Version 1.7.1: siggenes can also handle ExpressionSet objects. Attaching package: 'siggenes' The following object(s) are masked from package:multtest : .mt.BLIM The following object(s) are masked from package:multtest : .mt.naNUM > data(golub) Warning message: file 'golub.RData' has magic number 'RDA1' Use of save versions prior to 2 is deprecated > # golub.cl contains the class labels. > golub.cl [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 > # Perform a SAM analysis for the two class unpaired case assuming > # unequal variances. > sam.out<-sam(golub,golub.cl,B=100,rand=123) > sam.out SAM Analysis for the Two-Class Unpaired Case Assuming Unequal Variances Delta p0 False Called FDR 1 0.1 0.5 2424.77 2739 0.44276 2 0.7 0.5 262.21 1248 0.10508 3 1.3 0.5 12.11 507 0.01195 4 1.8 0.5 0.74 210 0.00176 5 2.4 0.5 0.01 76 6.58e-05 6 3.0 0.5 0 15 0 7 3.6 0.5 0 5 0 8 4.1 0.5 0 2 0 9 4.7 0.5 0 2 0 10 5.3 0.5 0 0 0 The software packages I'm using are: > sessionInfo() R version 2.4.0 Patched (2006-10-04 r39581) x86_64-unknown-linux-gnu locale: LC_CTYPE=en_US;LC_NUMERIC=C;LC_TIME=en_US;LC_COLLATE=en_US;LC_MONETARY =en_US;LC_MESSAGES=en_US;LC_PAPER=en_US;LC_NAME=C;LC_ADDRESS=C;LC_TELE PHONE=C;LC_MEASUREMENT=en_US;LC_IDENTIFICATION=C attached base packages: [1] "splines" "tools" "methods" "stats" "graphics" "grDevices" [7] "utils" "datasets" "base" other attached packages: siggenes multtest survival Biobase "1.8.0" "1.11.2" "2.29" "1.12.2" I found R 2.1.1 installed on a Windows laptop I have. The results of sam.out are nearly identical (the formatting in the R 2.1.1 version did not use scientific notation) to those presented above. Unfortunately I do not have access to the laptop at this instant, so cannot provide the detail above. The installed software was, I think, siggenes 1.2.17, so not exactly comparable to you. There are some suspicious items in the above, particularly the warning about 'Use of save versions prior to 2 is deprecated' and perhaps also the objects being 'masked'. It is hard to know whether these are worth pursuing as a source of your problems. How does your output differ? siggenes seems to be actively and conscientiously maintained, so a good source for help, if the mailing list fails you, might be the package author; try > library(help=siggenes) for contact information. Hope that helps, Martin "wzhao" <wzhao at="" mednet.ucla.edu=""> writes: > Hi, > > I am using SAM (from siggenes_1.2.11 package) method to select genes from > a microarray data set. After installing the latest R2.4.0 on my computer, to > my surprise the results are totally different from that calculated using > R2.1.1. Even the example code doesn't work the same way under these two > versions of R. Does anybody know what is going on? Thanks for any > suggestions. > > > > > > library(multtest) > data(golub) > # golub.cl contains the class labels. > golub.cl > # Perform a SAM analysis for the two class unpaired case assuming > # unequal variances. > sam.out<-sam(golub,golub.cl,B=100,rand=123) > sam.out > # Obtain the Delta plots for the default set of Deltas > plot(sam.out) > > > > > > Best > > Wei Zhao > Postdoctoral Research Fellow > Department of Human Genetics and Biostatistics > University of California, Los Angeles > Gonda Neuroscience and Genetics Research Center > 695 Charles E. Young Drive South, Box 708822 > Los Angeles, CA 90095-7088, USA > > Email: wzhao at mednet.ucla.edu > Tel:(310)8251677 > > > ---------------------------------------------------------- > IMPORTANT WARNING: This email (and any attachments) is only intended for the use of the person or entity to which it is addressed, and may contain information that is privileged and confidential. You, the recipient, are obligated to maintain it in a safe, secure and confidential manner. Unauthorized redisclosure or failure to maintain confidentiality may subject you to federal and state penalties. If you are not the recipient, please immediately notify us by return email, and delete this message from your computer. > ---------------------------------------------------------- > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting- guide.html > and provide commented, minimal, self-contained, reproducible code. -- Martin T. Morgan Bioconductor / Computational Biology http://bioconductor.org
Microarray Genetics Survival multtest siggenes safe Microarray Genetics Survival multtest • 933 views
ADD COMMENT
0
Entering edit mode
@holger-schwender-344
Last seen 8.3 years ago
Hi Wei Zhao, I have also taken a look on the results of sam for R 2.1.1 and 2.4, and they are exactly the same, and there are actually no reason -- at least in the sam code -- why the results might/should differ. Did you set the random number generator, i.e. rand, to the same value? I, however, do not get the warning Martin gets. This was once a warning in the R check of siggenes which then suddenly disappeared after I have asked Seth what this warning means. Best, Holger -------- Original-Nachricht -------- Datum: Mon, 20 Nov 2006 12:59:53 -0800 Von: Martin Morgan <mtmorgan at="" fhcrc.org=""> An: "wzhao" <wzhao at="" mednet.ucla.edu=""> Betreff: Re: [BioC] [R] Why SAM has totally diffent results in R2.1.1 and R2.4.0 > Wei Zhao > > This is not my software package and I do not use it regularly, but the > following might help in getting help from others: > > * Direct any follow-up to the bioconductor mailing list > > * Make sure that you are using a coherent set of current software. > > * Be more precise than 'totally different'. > > * Clearly indicate the software you are using. > > Following my own advice, I tried > > > update.packages() > > to update the installed packages; I might have done > > > source("http://bioconductor.org/biocLite.R") > > biocLite() > > on a new installation. I then ran your suggested code (with > library(siggenes) rather than multtest), with the following output: > > > library(siggenes) > Loading required package: multtest > Loading required package: Biobase > Loading required package: tools > > Welcome to Bioconductor > > Vignettes contain introductory material. To view, type > 'openVignette()' or start with 'help(Biobase)'. For details > on reading vignettes, see the openVignette help page. > > Loading required package: survival > Loading required package: splines > > Changes in siggenes: > Version 1.2.x: Added sam2excel, sam2html, and help.sam. > Version 1.5.3: Now exprSet objects can also be used in the EBAM functions > for specifying 'data' and 'cl'. > Version 1.7.1: siggenes can also handle ExpressionSet objects. > > > Attaching package: 'siggenes' > > > The following object(s) are masked from package:multtest : > > .mt.BLIM > > The following object(s) are masked from package:multtest : > > .mt.naNUM > > data(golub) > Warning message: > file 'golub.RData' has magic number 'RDA1' > Use of save versions prior to 2 is deprecated > > # golub.cl contains the class labels. > > golub.cl > [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 > 1 1 1 > > # Perform a SAM analysis for the two class unpaired case assuming > > # unequal variances. > > sam.out<-sam(golub,golub.cl,B=100,rand=123) > > sam.out > SAM Analysis for the Two-Class Unpaired Case Assuming Unequal Variances > > Delta p0 False Called FDR > 1 0.1 0.5 2424.77 2739 0.44276 > 2 0.7 0.5 262.21 1248 0.10508 > 3 1.3 0.5 12.11 507 0.01195 > 4 1.8 0.5 0.74 210 0.00176 > 5 2.4 0.5 0.01 76 6.58e-05 > 6 3.0 0.5 0 15 0 > 7 3.6 0.5 0 5 0 > 8 4.1 0.5 0 2 0 > 9 4.7 0.5 0 2 0 > 10 5.3 0.5 0 0 0 > > The software packages I'm using are: > > > sessionInfo() > R version 2.4.0 Patched (2006-10-04 r39581) > x86_64-unknown-linux-gnu > > locale: > LC_CTYPE=en_US;LC_NUMERIC=C;LC_TIME=en_US;LC_COLLATE=en_US;LC_MONETA RY=en_US;LC_MESSAGES=en_US;LC_PAPER=en_US;LC_NAME=C;LC_ADDRESS=C;LC_TE LEPHONE=C;LC_MEASUREMENT=en_US;LC_IDENTIFICATION=C > > attached base packages: > [1] "splines" "tools" "methods" "stats" "graphics" > "grDevices" > [7] "utils" "datasets" "base" > > other attached packages: > siggenes multtest survival Biobase > "1.8.0" "1.11.2" "2.29" "1.12.2" > > I found R 2.1.1 installed on a Windows laptop I have. The results of > sam.out are nearly identical (the formatting in the R 2.1.1 version > did not use scientific notation) to those presented > above. Unfortunately I do not have access to the laptop at this > instant, so cannot provide the detail above. The installed software > was, I think, siggenes 1.2.17, so not exactly comparable to > you. > > There are some suspicious items in the above, particularly the warning > about 'Use of save versions prior to 2 is deprecated' and perhaps also > the objects being 'masked'. It is hard to know whether these are worth > pursuing as a source of your problems. How does your output differ? > > siggenes seems to be actively and conscientiously maintained, so a good > source for help, if the mailing list fails you, might be the package > author; try > > > library(help=siggenes) > > for contact information. > > Hope that helps, > > Martin > > "wzhao" <wzhao at="" mednet.ucla.edu=""> writes: > > > Hi, > > > > I am using SAM (from siggenes_1.2.11 package) method to select genes > from > > a microarray data set. After installing the latest R2.4.0 on my > computer, to > > my surprise the results are totally different from that calculated using > > R2.1.1. Even the example code doesn't work the same way under these two > > versions of R. Does anybody know what is going on? Thanks for any > > suggestions. > > > > > > > > > > > > library(multtest) > > data(golub) > > # golub.cl contains the class labels. > > golub.cl > > # Perform a SAM analysis for the two class unpaired case assuming > > # unequal variances. > > sam.out<-sam(golub,golub.cl,B=100,rand=123) > > sam.out > > # Obtain the Delta plots for the default set of Deltas > > plot(sam.out) > > > > > > > > > > > > Best > > > > Wei Zhao > > Postdoctoral Research Fellow > > Department of Human Genetics and Biostatistics > > University of California, Los Angeles > > Gonda Neuroscience and Genetics Research Center > > 695 Charles E. Young Drive South, Box 708822 > > Los Angeles, CA 90095-7088, USA > > > > Email: wzhao at mednet.ucla.edu > > Tel:(310)8251677 > > > > > > ---------------------------------------------------------- > > IMPORTANT WARNING: This email (and any attachments) is only intended > for the use of the person or entity to which it is addressed, and may contain > information that is privileged and confidential. You, the recipient, are > obligated to maintain it in a safe, secure and confidential manner. > Unauthorized redisclosure or failure to maintain confidentiality may subject you > to federal and state penalties. If you are not the recipient, please > immediately notify us by return email, and delete this message from your > computer. > > ---------------------------------------------------------- > > > > [[alternative HTML version deleted]] > > > > ______________________________________________ > > R-help at stat.math.ethz.ch mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > > and provide commented, minimal, self-contained, reproducible code. > > -- > Martin T. Morgan > Bioconductor / Computational Biology > http://bioconductor.org > > _______________________________________________ > 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 --
ADD COMMENT

Login before adding your answer.

Traffic: 447 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6