Hi Giuliano, ComBat should work fine, just be cautious about outlier genes and samples, because these could influence the results. Outliers are usually less of a problem in microarray studies because there are so many genes, but in your case outliers could have more influence. To check this, look for abnormalities in the heatmap and also look for gross violations from normality in the prior plots generated by ComBat. Thanks! Evan On May 22, 2014, at 4:33 AM, Giuliano Stirparo <giulianostirparo at="" gmail.com=""> wrote: > Dear All > I am trying to use combat to perform batch correction of RT-PCR data. In particular I have around 40 genes and 59 samples. > The samples are divided in two categories (resistant on a particular treatment and sensitive) and they come from different batch (6 samples + 19 samples + 24 samples). > To perform batch correction I used the default parameter. I was wondering if it is possible use ComBat for this kind of analysis or if I have to try with different tools. > If you need more details please do not hesitate to contact me. > > Best > > Giuliano Stirparo

Dear All I am trying to use combat to perform batch correction of RT-PCR data. In particular I have around 40 genes and 59 samples. The samples are divided in two categories (resistant on a particular treatment and sensitive) and they come from different batch (6 samples + 19 samples + 24 samples). To perform batch correction I used the default parameter. I was wondering if it is possible use ComBat for this kind of analysis or if I have to try with different tools. If you need more details please do not hesitate to contact me. Best Giuliano Stirparo [[alternative HTML version deleted]]

Dear Evan, I have a similar question for a long time. For different platform (rtPCR vs. high-throughput array/sequencing), the number of genes vs. the number of samples, which one is a more important issue for ComBat, for example, here are two data I have: 1. a rtPCR data with 100 genes, but 2000 samples from 30 batches 2. a RNA-seq data with >25,000 genes, but only 14 samples from 2 different batches, is ComBat an optimal tool to correct batch effect comparing to removeBatchEffect() in edgeR? Many thanks, Shirley On Thu, May 22, 2014 at 7:07 AM, Johnson, William Evan <wej@bu.edu> wrote: > Hi Giuliano, > > ComBat should work fine, just be cautious about outlier genes and samples, > because these could influence the results. Outliers are usually less of a > problem in microarray studies because there are so many genes, but in your > case outliers could have more influence. To check this, look for > abnormalities in the heatmap and also look for gross violations from > normality in the prior plots generated by ComBat. > > Thanks! > > Evan > > > On May 22, 2014, at 4:33 AM, Giuliano Stirparo <giulianostirparo@gmail.com> > wrote: > > > Dear All > > I am trying to use combat to perform batch correction of RT-PCR data. In > particular I have around 40 genes and 59 samples. > > The samples are divided in two categories (resistant on a particular > treatment and sensitive) and they come from different batch (6 samples + 19 > samples + 24 samples). > > To perform batch correction I used the default parameter. I was > wondering if it is possible use ComBat for this kind of analysis or if I > have to try with different tools. > > If you need more details please do not hesitate to contact me. > > > > Best > > > > Giuliano Stirparo > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > [[alternative HTML version deleted]]

Thanks a lot! Inviato da iPhone > Il giorno 22/mag/2014, alle ore 13:07, "Johnson, William Evan" <wej at="" bu.edu=""> ha scritto: > > Hi Giuliano, > > ComBat should work fine, just be cautious about outlier genes and samples, because these could influence the results. Outliers are usually less of a problem in microarray studies because there are so many genes, but in your case outliers could have more influence. To check this, look for abnormalities in the heatmap and also look for gross violations from normality in the prior plots generated by ComBat. > > Thanks! > > Evan > > >> On May 22, 2014, at 4:33 AM, Giuliano Stirparo <giulianostirparo at="" gmail.com=""> wrote: >> >> Dear All >> I am trying to use combat to perform batch correction of RT-PCR data. In particular I have around 40 genes and 59 samples. >> The samples are divided in two categories (resistant on a particular treatment and sensitive) and they come from different batch (6 samples + 19 samples + 24 samples). >> To perform batch correction I used the default parameter. I was wondering if it is possible use ComBat for this kind of analysis or if I have to try with different tools. >> If you need more details please do not hesitate to contact me. >> >> Best >> >> Giuliano Stirparo >