Oh I may have misunderstood your question. There's no reason the average expression values (for any particular probe) should differ between limma or when you calculate them manually (assuming you aren't doing any additional normalization). You may have to provide code for somebody to be able to figure out the problem. Paul. On Mon, May 20, 2013 at 11:51 AM, Ekta Jain <ekta.jain at="" teri.res.in=""> wrote: > Dear Paul, > Thanks very much for your reply. My data is from rice and its an affymetrix array with 57,000 probes. The genes from limma show a high 'average expressio' for contrast of mock vs treated having 3 replicates in each group. When i do an average of the expression values of each of the three replicates from the gcrma normalised files, the value is much lower as compared to the one obtained from limma. > > I am unable to understand why is this so maybe becauase i do not know how limma calculates this average. > > Thank you, > Ekta > > > -----Paul Geeleher <paulgeeleher at="" gmail.com=""> wrote: ----- > > ======================= > To: Ekta Jain <ekta.jain at="" teri.res.in=""> > From: Paul Geeleher <paulgeeleher at="" gmail.com=""> > Date: 05/20/2013 07:37PM > cc: "bioconductor at r-project.org list" <bioconductor at="" r-project.org=""> > Subject: Re: [BioC] Average expression value from limma > ======================= > > Normally I think you'd expect that differentially expressed genes tend > to have higher average intensity. Consider that your array is > measuring approx 25,000 transcripts, this means that in your tissue > type many of these will not even be expressed, thus are highly > unlikely to be identified as differentially expressed. These probes > will obviously have the lowest flourescence intensity levels, thus > skewing the overall average flourescencce intensity level towards > zero. > > On some platforms it is possible to identify and remove these probes, > for example if you're using Affymetrix Exon arrays you can apply the > DABG algorithm. > > You'd probably still want to provide some info on how big the > difference is and what the array platform is though. > > Paul. > > > > On Mon, May 20, 2013 at 6:15 AM, Ekta Jain <ekta.jain at="" teri.res.in=""> wrote: >> >> Dear Bioconductor Mailing List, >> I have a microarray data which i normalised using 'gcrma' followed by limma >> for Mock vs Treated, 3 replicates in each group. The 'topTable' returns an >> average expression value for each probeset. How is this exactly calculated? >> >> Also i compared this with the average expression of each probeset for all >> samples, such as >> (Embedded image moved to file: pic17982.gif) >> >> The 'Avg.Expr.from.limma' has a very high value as compared to the ones >> obtained from the normalised data. Is there anything wrong? >> >> Much appreciate any comments please. >> >> Regards, >> Ekta Jain >> Research Analyst >> Biotechnology and Bio-resources Division >> The Energy and Resources Institute, India Habitat Centre >> Lodhi Road, New Delhi - 110033 >> #09958818853 >> ekta.jain at teri.res.in >> >> >> ------------------------------------------------------------------- ----------------------------------------- >> >> Disclaimer: >> >> The information contained in this e-mail is intended for the person or entity >> to which it is addressed, and it may contain confidential and/or privileged >> material. Any review or other use of this mail or taking any action based on it >> by persons or entities other than the intended recipient is strictly prohibited. >> If you receive this e-mail by mistake, please contact the sender, and delete all >> copies of this mail.This e-mail has been scanned and verified by McAfee SaaS >> Email Security, formerly MX Logic. >> _______________________________________________ >> 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 > > > > -- > Dr. Paul Geeleher, PhD (Bioinformatics) > Section of Hematology-Oncology > Department of Medicine > The University of Chicago > 900 E. 57th St., > KCBD, Room 7144 > Chicago, IL 60637 > -- > www.bioinformaticstutorials.com > > > > > -------------------------------------------------------------------- ---------------------------------------- > > Disclaimer: > > The information contained in this e-mail is intended for the person or entity > to which it is addressed, and it may contain confidential and/or privileged > material. Any review or other use of this mail or taking any action based on it > by persons or entities other than the intended recipient is strictly prohibited. > If you receive this e-mail by mistake, please contact the sender, and delete all > copies of this mail.This e-mail has been scanned and verified by McAfee SaaS > Email Security, formerly MX Logic. -- Dr. Paul Geeleher, PhD (Bioinformatics) Section of Hematology-Oncology Department of Medicine The University of Chicago 900 E. 57th St., KCBD, Room 7144 Chicago, IL 60637 -- www.bioinformaticstutorials.com