Question: question about p-values in maSigPro
0
10.1 years ago by
Dick Beyer1.4k
Dick Beyer1.4k wrote:
Hello Ana, I've been using the maSigPro package for some time-course data and I was wondering about how to interpret the output of p.vector(). I have two groups, young and old, 5 time points for each, 0,8,16,20,32, and no replicates. I'm trying, for example, degree=2. My design matrix is: design $dis tim tim2 young.0 0 0 young.8 8 64 young.16 16 256 young.20 20 400 young.32 32 1024 old.0 0 0 old.8 8 64 old.16 16 256 old.20 20 400 old.32 32 1024$groups.vector [1] "age" "age" $edesign tim repl age young.0 0 1 0 young.8 8 1 0 young.16 16 1 0 young.20 20 1 0 young.32 32 1 0 old.0 0 1 1 old.8 8 1 1 old.16 16 1 1 old.20 20 1 1 old.32 32 1 1 These are affy mouse gene ST arrays. In order to understand how to interpret the p.vector() p-values, I made a few genes have identical young and old expression values, for example, for one probe: age.time log2 expr young.0 5.969643 young.8 4.260117 young.16 4.650527 young.20 4.441443 young.32 4.955249 old.0 5.969643 old.8 4.260117 old.16 4.650527 old.20 4.441443 old.32 4.955249 What puzzles me is the p-value from the p.vector() fit is 0.006862036. I expected it to be about 1. I must be doing something wrong. My call to p.vector() is: fit <- p.vector(myData, design, Q = 1, min.obs = 3) where myData has 10 columns. sessionInfo() R version 2.9.0 (2009-04-17) i386-pc-mingw32 locale: 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] maSigPro_1.16.0 Biobase_2.4.0 loaded via a namespace (and not attached): [1] limma_2.18.0 Mfuzz_2.2.0 I'd appreciate any help or suggestions you could give me. Thanks very much, Dick ********************************************************************** ********* Richard P. Beyer, Ph.D. University of Washington Tel.:(206) 616 7378 Env. & Occ. Health Sci. , Box 354695 Fax: (206) 685 4696 4225 Roosevelt Way NE, # 100 Seattle, WA 98105-6099 http://depts.washington.edu/ceeh/ServiceCores/FC5/FC5.html http://staff.washington.edu/~dbeyer affy masigpro • 591 views ADD COMMENTlink modified 10.1 years ago • written 10.1 years ago by Dick Beyer1.4k Answer: question about p-values in maSigPro 0 10.1 years ago by Ana Conesa340 Ana Conesa340 wrote: Dear Dick I see that your edesign matrix is not well defined. It should be like: Time Replicates Young Old young.0 0 1 1 0 young.8 8 2 1 0 young.16 16 3 1 0 young.20 20 4 1 0 young.32 32 5 1 0 old.0 0 6 0 1 old.8 8 7 0 1 old.16 16 8 0 1 old.20 20 9 0 1 old.32 32 10 0 1 Try this and tell me how it goes now with the analysis Regards Ana On Thursday 16 July 2009 00:08:10 Dick Beyer wrote: > Hello Ana, > > I've been using the maSigPro package for some time-course data and I was > wondering about how to interpret the output of p.vector(). > > I have two groups, young and old, 5 time points for each, 0,8,16,20,32, and > no replicates. I'm trying, for example, degree=2. My design matrix is: > > design >$dis > tim tim2 > young.0 0 0 > young.8 8 64 > young.16 16 256 > young.20 20 400 > young.32 32 1024 > old.0 0 0 > old.8 8 64 > old.16 16 256 > old.20 20 400 > old.32 32 1024 > > $groups.vector > [1] "age" "age" > >$edesign > tim repl age > young.0 0 1 0 > young.8 8 1 0 > young.16 16 1 0 > young.20 20 1 0 > young.32 32 1 0 > old.0 0 1 1 > old.8 8 1 1 > old.16 16 1 1 > old.20 20 1 1 > old.32 32 1 1 > > These are affy mouse gene ST arrays. In order to understand how to > interpret the p.vector() p-values, I made a few genes have identical young > and old expression values, for example, for one probe: > > age.time log2 expr > young.0 5.969643 > young.8 4.260117 > young.16 4.650527 > young.20 4.441443 > young.32 4.955249 > old.0 5.969643 > old.8 4.260117 > old.16 4.650527 > old.20 4.441443 > old.32 4.955249 > > What puzzles me is the p-value from the p.vector() fit is 0.006862036. I > expected it to be about 1. I must be doing something wrong. > > My call to p.vector() is: > > fit <- p.vector(myData, design, Q = 1, min.obs = 3) > > where myData has 10 columns. > > sessionInfo() > R version 2.9.0 (2009-04-17) > i386-pc-mingw32 > > locale: > 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] maSigPro_1.16.0 Biobase_2.4.0 > > loaded via a namespace (and not attached): > [1] limma_2.18.0 Mfuzz_2.2.0 > > I'd appreciate any help or suggestions you could give me. > > Thanks very much, > Dick > ******************************************************************** ******* >**** Richard P. Beyer, Ph.D. University of Washington > Tel.:(206) 616 7378 Env. & Occ. Health Sci. , Box 354695 > Fax: (206) 685 4696 4225 Roosevelt Way NE, # 100 > Seattle, WA 98105-6099 > http://depts.washington.edu/ceeh/ServiceCores/FC5/FC5.html > http://staff.washington.edu/~dbeyer > > _______________________________________________ > 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 -- Ana Conesa Bioinformatics and Genomics Department Centro de Investigaciones Principe Felipe Avda. Autopista Saler 16, 46012 Valencia, Spain Phone: +34 96 328 96 80 Fax: +34 96 328 97 01 http://bioinfo.cipf.es/aconesa http://www.blast2go.org ========================================== FIRST INTERNATIONAL COURSE IN AUTOMATED FUNCTIONAL ANNOTATION AND DATA MINING Valencia/Orlando, September/October 2009 http://bioinfo.cipf.es/blast2gocourse
0
10.1 years ago by
Dick Beyer1.4k
Dick Beyer1.4k wrote:
Hi Ana, Thanks very much for the help. All is well now. Cheers, Dick ------------------------------------------ Message: 12 Date: Thu, 16 Jul 2009 11:54:04 +0200 From: Ana Conesa <aconesa@cipf.es> Subject: Re: [BioC] question about p-values in maSigPro To: bioconductor at stat.math.ethz.ch Message-ID: <200907161154.04821.aconesa at cipf.es> Content-Type: text/plain; charset="iso-8859-1" Dear Dick I see that your edesign matrix is not well defined. It should be like: Time Replicates Young Old young.0 0 1 1 0 young.8 8 2 1 0 young.16 16 3 1 0 young.20 20 4 1 0 young.32 32 5 1 0 old.0 0 6 0 1 old.8 8 7 0 1 old.16 16 8 0 1 old.20 20 9 0 1 old.32 32 10 0 1 Try this and tell me how it goes now with the analysis Regards Ana On Thursday 16 July 2009 00:08:10 Dick Beyer wrote: > Hello Ana, > > I've been using the maSigPro package for some time-course data and I was > wondering about how to interpret the output of p.vector(). > > I have two groups, young and old, 5 time points for each, 0,8,16,20,32, and > no replicates. I'm trying, for example, degree=2. My design matrix is: > > design > $dis > tim tim2 > young.0 0 0 > young.8 8 64 > young.16 16 256 > young.20 20 400 > young.32 32 1024 > old.0 0 0 > old.8 8 64 > old.16 16 256 > old.20 20 400 > old.32 32 1024 > >$groups.vector > [1] "age" "age" > > \$edesign > tim repl age > young.0 0 1 0 > young.8 8 1 0 > young.16 16 1 0 > young.20 20 1 0 > young.32 32 1 0 > old.0 0 1 1 > old.8 8 1 1 > old.16 16 1 1 > old.20 20 1 1 > old.32 32 1 1 > > These are affy mouse gene ST arrays. In order to understand how to > interpret the p.vector() p-values, I made a few genes have identical young > and old expression values, for example, for one probe: > > age.time log2 expr > young.0 5.969643 > young.8 4.260117 > young.16 4.650527 > young.20 4.441443 > young.32 4.955249 > old.0 5.969643 > old.8 4.260117 > old.16 4.650527 > old.20 4.441443 > old.32 4.955249 > > What puzzles me is the p-value from the p.vector() fit is 0.006862036. I > expected it to be about 1. I must be doing something wrong. > > My call to p.vector() is: > > fit <- p.vector(myData, design, Q = 1, min.obs = 3) > > where myData has 10 columns. > > sessionInfo() > R version 2.9.0 (2009-04-17) > i386-pc-mingw32 > > locale: > 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] maSigPro_1.16.0 Biobase_2.4.0 > > loaded via a namespace (and not attached): > [1] limma_2.18.0 Mfuzz_2.2.0 > > I'd appreciate any help or suggestions you could give me. > > Thanks very much, > Dick > ******************************************************************** ******* > **** Richard P. Beyer, Ph.D. University of Washington > Tel.:(206) 616 7378 Env. & Occ. Health Sci. , Box 354695 > Fax: (206) 685 4696 4225 Roosevelt Way NE, # 100 > Seattle, WA 98105-6099 > http://depts.washington.edu/ceeh/ServiceCores/FC5/FC5.html > http://staff.washington.edu/~dbeyer > > _______________________________________________ > 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 -- Ana Conesa Bioinformatics and Genomics Department Centro de Investigaciones Principe Felipe Avda. Autopista Saler 16, 46012 Valencia, Spain Phone: +34 96 328 96 80 Fax: +34 96 328 97 01 http://bioinfo.cipf.es/aconesa http://www.blast2go.org ========================================== ********************************************************************** ********* Richard P. Beyer, Ph.D. University of Washington Tel.:(206) 616 7378 Env. & Occ. Health Sci. , Box 354695 Fax: (206) 685 4696 4225 Roosevelt Way NE, # 100 Seattle, WA 98105-6099 http://depts.washington.edu/ceeh/ServiceCores/FC5/FC5.html http://staff.washington.edu/~dbeyer