Considering how little we actually know about the specific proteins coded by DNA in a cell, I don't think it's particularly wise to go around unilaterally declaring all these gene deserts.
I wouldn't be surprised to find that these deserts, much like the tonsils and appendix, aren't as vestigial and unused as biologists currently maintain they are.
Interestingly enough, an evolutionary paradigm led to the wrong conclusions about the appendix and tonsils being vestigial.
The absence of Open Reading Frames is good evidence for a "desert".
Gene sequencing technology has found many genes that were previously unknown by examining for open reading frames. This technology also allows us to sequence proteins, which was a real arduous task in the old days (sigh... I guess I am getting to the point where I belong with the dinosaurs).