I have a gene expression profile and miRNA expression profile for just normal samples.
I would like to test the relation between miRNAs and genes, however I want to do such analysis for just IMPORTANT genes (I am not sure if IMPORTANT genes is good choice for my goal). How can I find important genes in a gene expression profile while there is just one category of samples? Any thought?
Importance is a subjective not a scientific category. What is important depends on what you want to know and on your perspective. To put it differently, all genes might be important for the organism, otherwise they wouldn't be there. There are essential genes, but that is hardly a category you can find out using your design. Given a single sample there is not much you can find out about anything, except rank genes by absolute expression.
You might wish to correlate miRNA expression with their target genes, however if every sample is from the same condition, you might mostly try to detect pattern in noise. Also, post-transcriptional regulation is hard to detect by RNA-seq.
Next time, you should formulate your experimental question before doing the experiment and then design your experiment so it is appropriate to get an answer to your question.
