Yes, we are using Lumi VST to do the data transformation at this stage. Good luck, Wei affy snp wrote: > Thanks Wei. Fairly enough. I guess I will run to our microarray facility > tomorrow to request the slides re-scanned. BTW, I guess you are using > Lumi or Beadarray for normalization step? > > Regards, > Allen > > On Nov 11, 2007 5:28 PM, Wei Shi <shi at="" wehi.edu.au=""> wrote: > >> Dear Allen: >> >> We always work on the data set which are not background corrected >> and normalized. There are no negative expression values in this data set. >> >> Cheers, >> Wei >> >> >> affy snp wrote: >> >>> Dear Wei, >>> >>> Thanks a lot for the notes. We have also figured out by the help >>> of Illumina experts that the negative expression values were generated >>> by background subtraction. >>> >>> Our facility used the BeadStudio to do background subtraction and >>> normalization steps. The only concern I have is that we often see, >>> for a probset, it perhaps has positive value for control sample while >>> negative value for the test sample, or the opposite. For this scenario, >>> I do not know how to deal with. >>> >>> Have you also done the analysis starting with the dataset1 you mentioned? >>> Will that make a difference? >>> >>> We will also obtain the raw dataset without background subtraction and >>> see how it will look like. >>> >>> Thanks once again! >>> >>> Allen >>> >>> On Nov 11, 2007 7:13 AM, Wei Shi <shi at="" wehi.edu.au=""> wrote: >>> >>> >>>> You will get negative expression values if your data are global background >>>> corrected. Our microarray facility core creates two types of datasets: (1) >>>> Not background corrected and normalized and (2) background corrected and >>>> normalized. Background values can be seen in the probe control profile >>>> output from BeadStudio. >>>> >>>> Wei >>>> >>>> >>>> >>>> >>>>> On Nov 9, 2007 11:02 AM, affy snp <affysnp at="" gmail.com=""> wrote: >>>>> >>>>> >>>>>> Hi Sean, >>>>>> >>>>>> Thanks a lot for the suggestion. I checked with our microarray >>>>>> facility folks and they told me that they did background >>>>>> subtraction and then normalized the data in BeadStudio. >>>>>> Should we also skip the background subtraction step? >>>>>> >>>>>> >>>>> The negative values are generated by the normalization step, I >>>>> believe. So, you can do background subtraction or not. The choice to >>>>> opt out of background correction is relatively new in BeadStudio, I >>>>> think, so I do not have a systematic answer as to whether or not it is >>>>> warranted to use BeadStudio background subtraction or not. Perhaps >>>>> others will have strong opinions one way or the other. >>>>> >>>>> Sean >>>>> >>>>> _______________________________________________ >>>>> 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 >>>>> >>>>> >>>>> >>>>