Entering edit mode
Hi,
I wanted to ask if DEseq2 can handle zeros for some of the gene counts?
Is it better to add 1 to all the gene counts and that way not have zeros at all for better anaylsis?
Thank you very much,
Raya
Thank you very much.
Can I use DEseq2 if about 50% of the genes have zero counts in some of the replicates?
Thanks again,
Raya
I moved your post from an "Answer" to a "Comment".
The number of zeros itself is not an issue. You can use the plotSparsity() function to make a useful plot. What you don't want is to have the majority of genes with moderately high total counts (e.g. > 1000 sum of counts across samples) where a single sample makes up 99% of the total count. You should generally provide more information when you post a question, e.g. how many samples you have, what kind of comparison you want to make, is this typical bulk RNA-seq or some other experiment. Otherwise it's difficult to give more precise answers.
Thank you very much. Your answer was very helpful. I will try next time to ask more detailed questions.
Raya