Hi all 3 bio replicate and dye swap cDNA array experiment comparing 17day and 27day samples being analyzed with limma targets file: "Name" "FileNameCy3" "FileNameCy5" "Cy3" "Cy5" "1-Cy3-Ac-17d.txt" "1-Cy5-Ac-27d.txt" "17day" "27day" "2-Cy3-Ac-27d.txt" "2-Cy5-Ac-17d.txt" "27day" "17day" "3-Cy3-Ac-17d.txt" "3-Cy5-Ac-27d.txt" "17day" "27day" "4-Cy3-Ac27d.txt" "4-Cy5-Ac17d.txt" "27day" "17day" "5-Cy3-Ac17d.txt" "5-Cy5-Ac27d.txt" "17day" "27day" "6-Cy3-Ac27d.txt" "6-Cy5-Ac17d.txt" "27day" "17day" so after ... RG <- read.maimages(files, source="imagene") MA <- normalizeWithinArrays(RG) design=c(1,-1,1,-1,1,-1) fit <- lmFit(MA, design) . what direct are the fold changes I get out in does a positive FC mean up regulated in the 27day sample? Thanks (ps I would've found it useful if 'coefficients' were labelled logFC all the way through. But that might be due to my lack of stats knowledge!)

Hi Glyn, Your design matrix has the first array of the dye-swap pair as positive, which has Day27 in the Cy5. Because the M values you're testing are log2(R/G), yes a positive M value means upregulated in the Day27 samples. Two other suggestions: 1. If you only have 3 biological replicates, the dye-swaps are technical replicates. Your analysis is acting as if you had 6 biological replicates. See the limma vignette limmaUsersGuide() , section 8.2 to how to properly account for technical replicates. They have an example almost identical to your experiment (2 bio reps with dye-swaps). 2. It would be better to call your samples Day17 and Day27 - if you should ever have to use them for column names, R & limma can be funny about names that begin with a number. HTH, Jenny At 10:42 AM 7/17/2008, Glyn Bradley wrote: >Hi all > >3 bio replicate and dye swap cDNA array experiment comparing 17day and >27day samples being analyzed with limma >targets file: >"Name" "FileNameCy3" "FileNameCy5" "Cy3" "Cy5" > "1-Cy3-Ac-17d.txt" "1-Cy5-Ac-27d.txt" "17day" "27day" > "2-Cy3-Ac-27d.txt" "2-Cy5-Ac-17d.txt" "27day" "17day" > "3-Cy3-Ac-17d.txt" "3-Cy5-Ac-27d.txt" "17day" "27day" > "4-Cy3-Ac27d.txt" "4-Cy5-Ac17d.txt" "27day" "17day" > "5-Cy3-Ac17d.txt" "5-Cy5-Ac27d.txt" "17day" "27day" > "6-Cy3-Ac27d.txt" "6-Cy5-Ac17d.txt" "27day" "17day" > >so after >... >RG <- read.maimages(files, source="imagene") >MA <- normalizeWithinArrays(RG) >design=c(1,-1,1,-1,1,-1) >fit <- lmFit(MA, design) >. >what direct are the fold changes I get out in >does a positive FC mean up regulated in the 27day sample? > >Thanks > >(ps I would've found it useful if 'coefficients' were labelled logFC >all the way through. But that might be due to my lack of stats >knowledge!) > >_______________________________________________ >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 Jenny Drnevich, Ph.D. Functional Genomics Bioinformatics Specialist W.M. Keck Center for Comparative and Functional Genomics Roy J. Carver Biotechnology Center University of Illinois, Urbana-Champaign 330 ERML 1201 W. Gregory Dr. Urbana, IL 61801 USA ph: 217-244-7355 fax: 217-265-5066 e-mail: drnevich at illinois.edu

Quoting Glyn Bradley <glyn.bradley at="" googlemail.com="">: > Hi all > > 3 bio replicate and dye swap cDNA array experiment comparing 17day and > 27day samples being analyzed with limma > targets file: > "Name" "FileNameCy3" "FileNameCy5" "Cy3" "Cy5" > "1-Cy3-Ac-17d.txt" "1-Cy5-Ac-27d.txt" "17day" "27day" > "2-Cy3-Ac-27d.txt" "2-Cy5-Ac-17d.txt" "27day" "17day" > "3-Cy3-Ac-17d.txt" "3-Cy5-Ac-27d.txt" "17day" "27day" > "4-Cy3-Ac27d.txt" "4-Cy5-Ac17d.txt" "27day" "17day" > "5-Cy3-Ac17d.txt" "5-Cy5-Ac27d.txt" "17day" "27day" > "6-Cy3-Ac27d.txt" "6-Cy5-Ac17d.txt" "27day" "17day" > > so after > ... > RG <- read.maimages(files, source="imagene") > MA <- normalizeWithinArrays(RG) > design=c(1,-1,1,-1,1,-1) > fit <- lmFit(MA, design) > . > what direct are the fold changes I get out in > does a positive FC mean up regulated in the 27day sample? > > Thanks If in doubt, and you don't understand the design matrix, you can simply find a probe with a large positive M value, and then look at the raw intensities for that one. Then it will be obvious. Jose -- Dr. Jose I. de las Heras Email: J.delasHeras at ed.ac.uk The Wellcome Trust Centre for Cell Biology Phone: +44 (0)131 6513374 Institute for Cell & Molecular Biology Fax: +44 (0)131 6507360 Swann Building, Mayfield Road University of Edinburgh Edinburgh EH9 3JR UK -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.