Hi,
I have a dataset of shRNA-Seq library with multiple shRNA for many different genes. I have done the analysi on each of the 187 shRNA molecules. Now i would like to combine the analysis on the gene level. Is it possible, just to merge the results for all of the shRNA molecules of each gene, by calculating the average of each group?
Is there another way to run the analysis on gene level?
Does it make sense to use roast here?
The assumption in roast, if I understood it correctly, is that if 25%-50% of the genes in a specific gene set are DE, the whole is being considered as DE. In my case I have "only" up to three-four shRNA molecules in a gene set. This means, that as soon one of the shRNA molecules were found to be DE, the whole set will automatically be identified as DE. Is this correct?
Is there a way to make it more stringent (does it make sense to do it so)?
Thanks
