Michael Love Hi. I am following one of your courses about Introduction to Bioconductor. I have a question regarding the code that is given to answer an exercise. I am required to perform the following code but on a different GO:
library(GEOquery) e = getGEO("GSE34313")[] e$condition = e$characteristics_ch1.2 levels(e$condition) = c("dex24","dex4","control") table(e$condition) names(fData(e)) fData(e)$GO_ID[1:4] lvls = c("control", "dex4") es = e[,e$condition %in% lvls] es$condition = factor(es$condition, levels=lvls) library(limma) library(qvalue) design = model.matrix(~ es$condition) **fit = lmFit(es, design=design)** fit = eBayes(fit) topTable(fit)[,c(6,7,18,22)] set.seed(1) idx = grep("GO:0006955", fData(es)$GO_ID) length(idx) **r1 = roast(es, idx, design)** r1
However, when I type fit = lmFit(es, design=design) it shows "Coefficients not estimable: (Intercept) es$conditiondex4 Error in lm.fit(design, t(M)) : 0 (non-NA) cases". I wonder why is that. Since fit is not functioning properly, the roast function is not working properly as well. It shows: "Error in .lmEffects(y = y, design = design, contrast = contrast, array.weights = Dots$array.weights, : No residual degrees of freedom"
In the video of the course, there seems to be no problem with the code, but now there is. If I could get a better insight about what is happening, and how I could correct these functions, I would be thankful. Best regards!