Do we need to correct GC-content bias and/or length bias in RNA-Seq level 3 from TCGA?

Question

[Na Sed]

I am working on interactions between miRNAs and mRNAs using their RNA-Seq files obtained from TCGA database in level 3.0 format. Now, my professor has asked me about correction of GC content bias and length bias in pre-processing steps.

What is your idea about her questions? Do we need to do such corrections for level 3.0 RNA-Seq data from TCGA?

Thanks