I conducted a time-course with 5 time points and two samples (three replicates for each time point).I have a dds object with 30 columns( one for each replicate) that I have vst-transformed. I have my columns as samples and my rows as transcripts. I have about 8000 DEGs and would like to make a heatmap with all of them to look at genome-wide clusters before I drill down into time-specific samples. I have been using the following function but can't get my heatmap to assemble (R keeps aborting). I assume it is because there are too many DEGs but I was wondering if anybody knew if making the heatmap was feasible.
I have been using the following code (assuming vsd is my vst-transformed dds object with the 8000 DEGs):
heatmap(vsd, Rowv=TRUE, Colv = FALSE, scale = "row")
Kevin's tutorial should be useful: https://github.com/kevinblighe/E-MTAB-6141 he uses ComplexHeatmap package, another option is heatmap.2
Thank you. I ended up getting to the root of the problem and I can produce the heatmap now. However, I will read these tutorials because ComplexHeatmap looks rather impressive. Thanks!
Log in to answer this question.
