Hi John In made4, For correspondence analysis (COA, the default option in made4) the total $eig will be equivalent to the total chi-sq of the table. data.coa <- ord(eSetdata, type = "coa") To get the % of variance explained by each axis. > data.coa$ord$eig * 100/sum(data.coa$ord$eig) [1] 33.69 23.13 12.65 9.84 6.44 5.83 4.66 3.76 The cumulative variance is given by > cumsum(data.coa$ord$eig * 100/sum(data.coa$ord$eig)) [1] 33.7 56.8 69.5 79.3 85.7 91.6 96.2 100.0 Therefore almost 57% of the variance is captured by the first 2 components. For PCA data.pca <- ord(eSetdata, type = "pca") To get the % of variance explained by each axis. data.pca$ord$eig * 100 The cumulative variance is given by cumsum(data.coa$ord$eig * 100) Examples of these and more complete explanation are given on the website http://compbio.dfci.harvard.edu/courses/bioconductor/ with the vignette url http://compbio.dfci.harvard.edu/courses/bioconductor/MGED_Oct09.pdf best regards Aedin Message: 7 Date: Mon, 18 Oct 2010 11:53:40 +0200 From: "Gerhard Thallinger" <gerhard.thallinger@tugraz.at> To: "'John Antony Gaspar'" <gasparj@uni-koeln.de> Cc: bioconductor@stat.math.ethz.ch Subject: Re: [BioC] Percentage of Variance to be included at 3D graph of Principal Component Analysis (PCA)- made4 package Message-ID: <a1317a5fae314443af5958a387e95419@agilix> Content-Type: text/plain; charset="us-ascii" Dear John, > > Please help me out with this regard. 'prcomp' provides Standard > > deviation, Proportion of variance and Cumulative Proportion. How to > > use them in calculating the percentage of variance at all three > > PC level. > The percentage of variance explained can be calculated as follows: pca <- prcomp(USArrests, scale = TRUE) pc1 <- round(pca$sdev[1]2/sum(pca$sdev2)*100,2) pc2 <- round(pca$sdev[2]2/sum(pca$sdev2)*100,2) etc ... Hth Gerhard ---------------------------------------------------------------------- -- Dr. Gerhard Thallinger E-mail: Gerhard.Thallinger@tugraz.at Institute for Genomics and Bioinformatics Web: http://genome.tugraz.at Graz University of Technology Tel: +43 316 873 5343 Petersgasse 14/V Fax: +43 316 873 105343 8010 Graz, Austria Map: http://genome.tugraz.at/Loc.html -- Aedín Culhane, Computational Biology and Functional Genomics Harvard School of Public Health, Dana-Farber Cancer Institute 44 Binney Street, SM822C Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute Boston, MA 02115 USA Phone: +1 (617) 632 2468 Fax: +1 (617) 582 7760 Email: aedin@jimmy.harvard.edu Web URL: http://www.hsph.harvard.edu/research/aedin-culhane/ [[alternative HTML version deleted]]