Thanks. Yolande. On Sun, Nov 14, 2010 at 5:34 PM, Mark Cowley <m.cowley at="" garvan.org.au=""> wrote: > Hi Yolande, > Are your protein counts on the log, or linear scale? If log, use difference, if linear, use fraction, > HTH > Mark > > On 13/11/2010, at 3:07 AM, Yolande Tra wrote: > >> Hi list members, >> >> May I have your expert opinion on the following. >> I am looking for protein differentially expressed due to treatment of >> a certain disease. Protein counts were measured before and after >> treatment was given. Some patients respond to the treatment >> (responders) and got better and some patients did not respond to the >> treatment (nonresponders). The goal is to find which protein is >> responsible for the positive response, so really looking at group >> factor. >> I have used two quantities to describe the change of these proteins: >> ratio (after/before) and difference (after-before). After computing >> these quantities, I considered the design as two independent groups >> (responders versus nonresponders). >> I used rankproduct procedure to detect the proteins of interest. I >> obtained different lists of protein differentially expressed which >> means that using different quantities afftect the results. >> >> My question is which quantity (fraction or difference) is more >> sensitive to a change in this situation? >> >> Thank you for any suggestion and any insight in this question, >> Yolande >> >> _______________________________________________ >> 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 > >