Wei Shi <shi at="" ...=""> writes: > > Hi Prasad: > > I had a close look at one of the datasets: > http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? view=data&acc=GSM614020&id=15413&db=GeoDb_blob51 > . The header information in this page is: > > #ID_REF = > #VALUE = Quantile, subtract background > #1690411024_A.Detection Pval = > > It looks like they performed quantile normalization to the data but also subtracted probe intensities by > the background intensity. The negative values in the table seemed to result from the "subtract > background" step. However, it is unclear how the data was analyzed at all. > > I would suggest you to contact the authors who generated these data. This might be the easiest and quickest > way to get around this issue and get the data you want. > > Hope this helps. > > Cheers, > Wei Dear Wei, Thank you very much for the information. Do you think the data is log2 transformed? I can't transform because of negative values (My original problem). Can you please answer my question I posted today with subject "limma complex contrast matrix" on 011-04-07 17:05:59 GMT Prasad

Hi Prasad: Have a look at this page: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM614020 . It is clear now they used BeadStudio for the analysis. So probe intensities were firstly subtracted by mean background intensity and then quantile normalized. The data was not log transformed because they are definitely not on the log scale. I am not sure if I can answer your previous post about the complex contrast matrix because I am not a statistician. But you can post it again to see if it can be answered by someone else. Cheers, Wei On Apr 8, 2011, at 2:15 PM, Prasad Siddavatam wrote: > > Wei Shi <shi at="" ...=""> writes: > >> >> Hi Prasad: >> >> I had a close look at one of the datasets: >> http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? > view=data&acc=GSM614020&id=15413&db=GeoDb_blob51 >> . The header information in this page is: >> >> #ID_REF = >> #VALUE = Quantile, subtract background >> #1690411024_A.Detection Pval = >> >> It looks like they performed quantile normalization to the data but also > subtracted probe intensities by >> the background intensity. The negative values in the table seemed to result > from the "subtract >> background" step. However, it is unclear how the data was analyzed at all. >> >> I would suggest you to contact the authors who generated these data. This > might be the easiest and quickest >> way to get around this issue and get the data you want. >> >> Hope this helps. >> >> Cheers, >> Wei > > > Dear Wei, > > Thank you very much for the information. Do you think the data is log2 > transformed? I can't transform because of negative values (My original problem). > > Can you please answer my question I posted today with subject "limma > complex contrast matrix" on 011-04-07 17:05:59 GMT > > > Prasad > > _______________________________________________ > 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 ______________________________________________________________________ The information in this email is confidential and intend...{{dropped:6}}