Hi everybody! I know that there are a lot of questions about how to adjust the RNA-Seq data if there is a batch effect in the samples, but after reading a lot of posts, questions and recommendations, I still have dobts on how to deal with my samples.
I have RNA-seq data from 2 different samples (control and treatment), which I want to compare and result in DEG list. For each sample, I have 3 replicates.
I have used edgeR to normalize the data (method=TMM), and after that I have performed the MDSplot and PCA plot. In both of them I see that there is a batch effect between the samples collected on the same day, as instead of seeing two separated groups between control and treatment, I see 3 groups with pairs of control_1 and treatment_1, control_2 and treatment_2 and control_3 and treatment_3. I see this in both MDSplot and PCAplot.
So I have been reading in biostars, and decided to perform batch effect correction in edgeR defining "batch" and unsing in in the design matrix for the DE analysis.
But to see if this adjusting of the batch effect works I have used "removeBatchEffect" function to then plot MDS and PCA and see if I obtain that separation. I think that for MDS plot more or less it is correct the separation, but again in PCA plot there is something with control_1 and treatment_1 that doesn't work even after the batch adjustment.
How can I solve this to be sure that when performing the DE analysis I am considering properly the samples and replicates, and finally have a good looking MDS and PCA plots?
Thanks in advance!!