request
2
0
Entering edit mode
@alexey-moskalev-6079
Last seen 9.6 years ago
I am using DeSeq package to produce Principal components biplot on variance stabilized data for my RNASeq data. I was wondering if you advice me how to know Proportion of Variance for the first and the second Principal components using DeSeq? [[alternative HTML version deleted]]
RNASeq DESeq RNASeq DESeq • 1.5k views
ADD COMMENT
0
Entering edit mode
@wolfgang-huber-3550
Last seen 3 months ago
EMBL European Molecular Biology Laborat…
Dear Alexey please type "plotPCA" into the R command line to see how the function computes the PCA, then have a look at the manual page of the functions "prcomp" and "screeplot". @all: I am not sure what would be a good user interface would be for modifying the "plotPCA" function so that it can return the 'pca' object for user inspection (such as desired by Alexey); currently it returns the 'trelliis' object as its return value. Best wishes Wolfgang On 6 Aug 2013, at 08:54, Alexey Moskalev <amoskalev at="" list.ru=""> wrote: > I am using DeSeq package to produce Principal components biplot on variance stabilized data for my RNASeq data. I was wondering if you advice me how to know Proportion of Variance for the first and the second Principal components using DeSeq? > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
ADD COMMENT
0
Entering edit mode
On 6 August 2013 16:32, Wolfgang Huber <whuber at="" embl.de=""> wrote: > @all: I am not sure what would be a good user interface would be for modifying the "plotPCA" function so that it can return the 'pca' object for user inspection (such as desired by Alexey); currently it returns the 'trelliis' object as its return value. I have a similar function (pRoloc::plot2D) that invisibly returns the prcomp(...)$x[, dims] matrix that used for plotting, where dims are the PCs requested by the user (default being 1:2). I also report the proportion of variance explained by these two components on the axes. Best wishes, Laurent > Best wishes > Wolfgang > > > > On 6 Aug 2013, at 08:54, Alexey Moskalev <amoskalev at="" list.ru=""> wrote: > >> I am using DeSeq package to produce Principal components biplot on variance stabilized data for my RNASeq data. I was wondering if you advice me how to know Proportion of Variance for the first and the second Principal components using DeSeq? >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor at r-project.org >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- Laurent Gatto - http://proteome.sysbiol.cam.ac.uk/lgatto/ Cambridge Centre for Proteomics - http://www.bio.cam.ac.uk/proteomics Using R/Bioconductor for proteomics data analysis - http://lgatto.github.io/RforProteomics/
ADD REPLY
0
Entering edit mode
Dear Laurent in pRoloc::plot2D the plot is a side effect (via graphics::plot), and therefore you are free to return something else; while in the function discussed below the plot (a 'trellis' object) is the return value, which then usually is rendered via 'print.trellis'. (One could stick additional information like the PCA loadings and eigenvalues into the same (S3-)object, initially I thought this was ugly but maybe it's the way to go.) Best wishes Wolfgang On Aug 6, 2013, at 9:33 pm, Laurent Gatto <lg390 at="" cam.ac.uk=""> wrote: > On 6 August 2013 16:32, Wolfgang Huber <whuber at="" embl.de=""> wrote: >> @all: I am not sure what would be a good user interface would be for modifying the "plotPCA" function so that it can return the 'pca' object for user inspection (such as desired by Alexey); currently it returns the 'trelliis' object as its return value. > > I have a similar function (pRoloc::plot2D) that invisibly returns the > prcomp(...)$x[, dims] matrix that used for plotting, where dims are > the PCs requested by the user (default being 1:2). I also report the > proportion of variance explained by these two components on the axes. > > Best wishes, > > Laurent > >> Best wishes >> Wolfgang >> >> >> >> On 6 Aug 2013, at 08:54, Alexey Moskalev <amoskalev at="" list.ru=""> wrote: >> >>> I am using DeSeq package to produce Principal components biplot on variance stabilized data for my RNASeq data. I was wondering if you advice me how to know Proportion of Variance for the first and the second Principal components using DeSeq? >>> [[alternative HTML version deleted]] >>> >>> _______________________________________________ >>> Bioconductor mailing list >>> Bioconductor at r-project.org >>> https://stat.ethz.ch/mailman/listinfo/bioconductor >>> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor at r-project.org >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > > > > -- > Laurent Gatto > - http://proteome.sysbiol.cam.ac.uk/lgatto/ > Cambridge Centre for Proteomics > - http://www.bio.cam.ac.uk/proteomics > Using R/Bioconductor for proteomics data analysis > - http://lgatto.github.io/RforProteomics/
ADD REPLY
0
Entering edit mode
Hi, On Tue, Aug 6, 2013 at 12:49 PM, Wolfgang Huber <whuber at="" embl.de=""> wrote: > Dear Laurent > in pRoloc::plot2D the plot is a side effect (via graphics::plot), and therefore you are free to return something else; while in the function discussed below the plot (a 'trellis' object) is the return value, which then usually is rendered via 'print.trellis'. > > (One could stick additional information like the PCA loadings and eigenvalues into the same (S3-)object, initially I thought this was ugly but maybe it's the way to go.) Along these lines: is it considered "bad form" to just add a "pca" `attr`-ibute to the xyplot object you are returning? eg: plotPCA <- function(...) { ## ... out <- xyplot(PC2 ~ PC1, ...) attr(out, 'pca') <- pca invisible(out) ## or not invisible } -- Steve Lianoglou Computational Biologist Bioinformatics and Computational Biology Genentech
ADD REPLY
0
Entering edit mode
Dear Wolfgang! Greate, it works! Thank you so much! Alex Вторник, 6 августа 2013, 17:32 +02:00 от Wolfgang Huber <whuber@embl.de>: >Dear Alexey >please type "plotPCA" into the R command line to see how the function computes the PCA, then have a look at the manual page of the functions "prcomp" and "screeplot". > >@all: I am not sure what would be a good user interface would be for modifying the "plotPCA" function so that it can return the 'pca' object for user inspection (such as desired by Alexey); currently it returns the 'trelliis' object as its return value. > >Best wishes >Wolfgang > > > >On 6 Aug 2013, at 08:54, Alexey Moskalev < amoskalev@list.ru > wrote: > >> I am using DeSeq package to produce Principal components biplot on variance stabilized data for my RNASeq data. I was wondering if you advice me how to know Proportion of Variance for the first and the second Principal components using DeSeq? >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor@r-project.org >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > Sincerely, Dr. Alexey Moskalev Head of the Laboratory of Molecular Radiobiology and Gerontology Institute of Biology, Komi Science Center of RAS, Kommunisticheskaya St.28 167982, Syktyvkar Russia Blog: http://aging-genes.livejournal.com/ [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
@alexey-moskalev-6079
Last seen 9.6 years ago
Thanks a lot! Вторник, 6 августа 2013, 20:33 +01:00 от Laurent Gatto <lg390@cam.ac.uk>: >On 6 August 2013 16:32, Wolfgang Huber < whuber@embl.de > wrote: >> @all: I am not sure what would be a good user interface would be for modifying the "plotPCA" function so that it can return the 'pca' object for user inspection (such as desired by Alexey); currently it returns the 'trelliis' object as its return value. > >I have a similar function (pRoloc::plot2D) that invisibly returns the >prcomp(...)$x[, dims] matrix that used for plotting, where dims are >the PCs requested by the user (default being 1:2). I also report the >proportion of variance explained by these two components on the axes. > >Best wishes, > >Laurent > >> Best wishes >> Wolfgang >> >> >> >> On 6 Aug 2013, at 08:54, Alexey Moskalev < amoskalev@list.ru > wrote: >> >>> I am using DeSeq package to produce Principal components biplot on variance stabilized data for my RNASeq data. I was wondering if you advice me how to know Proportion of Variance for the first and the second Principal components using DeSeq? >>> [[alternative HTML version deleted]] >>> >>> _______________________________________________ >>> Bioconductor mailing list >>> Bioconductor@r-project.org >>> https://stat.ethz.ch/mailman/listinfo/bioconductor >>> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor@r-project.org >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > > > >-- >Laurent Gatto >- http://proteome.sysbiol.cam.ac.uk/lgatto/ >Cambridge Centre for Proteomics >- http://www.bio.cam.ac.uk/proteomics >Using R/Bioconductor for proteomics data analysis >- http://lgatto.github.io/RforProteomics/ Sincerely, Dr. Alexey Moskalev Head of the Laboratory of Molecular Radiobiology and Gerontology Institute of Biology, Komi Science Center of RAS, Kommunisticheskaya St.28 167982, Syktyvkar Russia Blog: http://aging-genes.livejournal.com/ [[alternative HTML version deleted]]
ADD COMMENT

Login before adding your answer.

Traffic: 730 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