Dear Xiaokuan, > Date: Wed, 11 Aug 2010 21:12:04 -0700 (PDT) > From: Xiaokuan Wei <weixiaokuan at="" yahoo.com=""> > To: bioconductor at stat.math.ethz.ch > Subject: [BioC] Long Time Series analysis > > Dear List, > > I?was using?maSigPro for time series analysis of microarray.?I > have about 12 > time points for a yeast data, and each time point only has two > replicates. I am > wondering if I can use quadratic model?to perform the analysis. If I > used higher > order model, will overfitting become a problem? I've done analysis of time-course experiment with seven points, I considered to use 3rd or 4th order polynomial. Eventually it turned out that 3rd order is enought, 4th order gives nothing new. I advise you to try with quadratic model, 3rd, 4th and maybe higer order polynomial too and to look if it gives different results, if not lower level polynomial is enought. > > I noticed that in?maSigPro tutorial, 4.4 mentioned "Long Time > Series". But it > has no detailed explaination. Does Ana or someone can share?some > examples I can > follow for such analysis? I've asked for longer than mentioned 5 time points series and I've been informed that in case of 7 time points I can use normal workflow. > > Any other packages I can use for a long time series analysis? You can look in "packages" section at the Bioconductor website. Hope it helps somehow. > Your help will be greatly appreciated. > > Thanks again. > > Regards, > Xiaokuan > Maciej Jo?czyk, MSc Department of Plant Molecular Ecophysiology Institute of Plant Experimental Biology Faculty of Biology, University of Warsaw 02-096 Warszawa, Miecznikowa 1 ___________________________________ NOCC, http://nocc.sourceforge.net

Thank you Maciej, your experience is really helpful. I'll try to add higher orders step by step. -Xiaokuan ________________________________ From: Maciej Jończyk <mjonczyk@biol.uw.edu.pl> Cc: Maciej Jończyk <mjonczyk@biol.uw.edu.pl> Sent: Wed, August 18, 2010 4:38:51 AM Subject: Re: [BioC] Long Time Series analysis Dear Xiaokuan, > Date: Wed, 11 Aug 2010 21:12:04 -0700 (PDT) > To: bioconductor@stat.math.ethz.ch > Subject: [BioC] Long Time Series analysis > > Dear List, > > I?was using?maSigPro for time series analysis of microarray.?I > have about 12 > time points for a yeast data, and each time point only has two > replicates. I am > wondering if I can use quadratic model?to perform the analysis. If I > used higher > order model, will overfitting become a problem? I've done analysis of time-course experiment with seven points, I considered to use 3rd or 4th order polynomial. Eventually it turned out that 3rd order is enought, 4th order gives nothing new. I advise you to try with quadratic model, 3rd, 4th and maybe higer order polynomial too and to look if it gives different results, if not lower level polynomial is enought. > > I noticed that in?maSigPro tutorial, 4.4 mentioned "Long Time > Series". But it > has no detailed explaination. Does Ana or someone can share?some > examples I can > follow for such analysis? I've asked for longer than mentioned 5 time points series and I've been informed that in case of 7 time points I can use normal workflow. > > Any other packages I can use for a long time series analysis? You can look in "packages" section at the Bioconductor website. Hope it helps somehow. > Your help will be greatly appreciated. > > Thanks again. > > Regards, > Xiaokuan > Maciej Jończyk, MSc Department of Plant Molecular Ecophysiology Institute of Plant Experimental Biology Faculty of Biology, University of Warsaw 02-096 Warszawa, Miecznikowa 1 ___________________________________ NOCC, http://nocc.sourceforge.net [[alternative HTML version deleted]]