Entering edit mode
Hello,
I did plotcounts shows normalized counts for a single gene. two groups contains two sample which are not close together actually they are not replicates indeed, but the same library sequenced on two lanes. so I would appreciate let me know about such cases how should I do? can I assign them in one group for differential gene expression as an next step?
can we do collapseReplicates for each cases, do not have any exclusion?
I'm sorry, I don't exactly follow the experimental setup. Can you list the samples you have and what comparisons you want to make with DESeq2?
about that case, I have 8 samples, "
Dnmt2_1ND", "Dnmt2_1NF", "Dnmt2_2ND", "Dnmt2_2NF", "Dnmt2_3HD", "Dnmt2_3HF", "Dnmt2_4HD", "Dnmt2_4HF
ND ,NF , HD and HF have same condition, but they are not tecnical replicates, actuelly their sample has been taken from two different mouse but with the same condition, I think we should call them biological replicates, isn't it? if so, which analysis is better to do for DGE? I want to campare Dnmt2_1ND vs Dnmt2_1NF and Dnmt2_2HD vs Dnmt2_2HF ....?
This isn’t possible with our software. It requires replicates to run DESeq().
You can compute vst() and look at the fold changes though.
Right, if I assign Dnmt2_1ND and Dnmt2_2ND as two biological replicates, may I use DESeq to make a compare with Dnmt2_1NF and Dnmt2_2NF together?
and for another samples which I had, I did DESeq2. I had D1, D3, R1 and R3. each of them has two replicates, D1_L001 and D1_L002 and same this for others. I got transcript quantification with Salmon (and quantified technical replications together), then import it to tximport. plotted with normalized counts each D1 and D3 were close together, and I did DGE (D (plus D1 and D3) vs R(R1 and R3) with DESeq. may I trust to these results?
From what I understand from your description, you want to make comparisons of individual samples, sometimes by combining other samples. I would not recommend this. It sounds arbitrary and you’re likely to trick yourself into a false positive result.
Excuse me, I realised that they are not technical replicates, they have same condition but from two separate samples, so I can't do DGE by Deseq2 bacause it needs replicates, what is your suggestion?