Hi All, I was wondering if anyone had any idea how one might put error bars on the log fold changes calculated by edgeR? I'd like to do this so I can show examples of differentially expressed genes on a bar plot, and show them in comparison to qPCR validation. I feel like one should be able to get the variance from the dispersion: CV^2 = 1/mu + BCV^2 = 1/mu + dispersion So: variance = mu + mu^2*dispersion ? This could either be used to show the variance for each of the two conditions, or used to calculate an error for the fold change. but I'm a bit stuck as to how to actually implement this. Any ideas appreciated, Yours, Ian [[alternative HTML version deleted]]