Hi, I am trying to get RNA sequences for my model to predict protein binding sites in RNA. I am trying to mimic the method of other groups and could not understand some procedures in their preparation. They got RNA sequences from ENCODE eCLIP result. Then found an overlap region of the RNA sequences with the corresponding reference gene. Why do we need the overlap region? instead of just taking the RNA sequence of the processed data file in ENCODE? Is this about eliminating the added tails in eCLIP pipeline? Thank you.