Dear all,

I am using deseq2 in order to calculate differential expression between different experimental conditions with two replicates for each. I understand deseq2 perform a normalization on the reads count in order to make the experiments comparable when different overall coverage is obtained. For this reason I can easily create an heatmap to visually compare the expression of treated and untreated samples. What about normalizing the all dataset? I have several experiments, collection times, plant variety and so on, and I have raw counts for each of them. I would like to perform a PCA on all the samples and for this reason I guess I should perform a normalization that include all the samples data. How can I do that? Should I create a DeseqDataSet object with all the raw counts? How would it look like in this case the coldata file?

Thanks a lot for the support!

Yes, you can create one large DeSeqDataSet. The coldata can include information from all the experiments.