I have performed a matched sample miRNA and mRNA profiling to identify the potential mechanisms that regulate aspects that are associated with cancer stem cell properties in my treated samples. I think of integrating both the miRNA and mRNA profiles, so I am doing the miRNA target prediction analysis at mirDB and TargetScan for now, as I thought of adding few more. My question is, for the target genes, what is the best way to compare these predicted target with my mRNA profiling data?

1. Do I compare with the overlapping predicted targets from both miRDB and TargetScan with my mRNA profiling data? or

2. Do I compare each predicted target from each tools with my mRNA results, then take the combination (union) of both of these?

Additional: What do you do if you have more than 3 target prediction tool and how do you compare the predicted targets with your mRNA results?

Thank you so much for your kind help.