Hi Faranak, Did it work for you the WGCNA with the normalized values? Did you find another possible solution? When I use the normalized values it gives me a weird scale free topology model fit. Thank you, Spanos [[alternative HTML version deleted]]

Hi Spanos, you can certainly use WGCNA for RNA-seq data. Two recommendations: 1. Filter out genes whose count is less than say 5 in more than say 80% of the samples. This gets rid of a lot of noise and gets rid of expression profiles for which correlation makes little sense. 2. Use a variance-stabilizing transformation, such as the one implemented in varianceStabilizingTransformation or rlogTransformation in the DESeq2. I have analyzed a few RNA-seq data sets and have had great results. Hope this helps, Peter On Thu, Mar 6, 2014 at 4:32 PM, spano spano <diadiktuo at="" outlook.com=""> wrote: > Hi Faranak, > > Did it work for you the WGCNA with the normalized values? Did you find another possible solution? > When I use the normalized values it gives me a weird scale free topology model fit. > > Thank you, > Spanos > > > [[alternative HTML version deleted]] > > _______________________________________________ > 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