Mutations-change conformation

Dear ZhangLab men,

recently I'm using your server to predict the murine form of my target (indolamine 2,3-dioxygenase, no X-ray available for the murine form). The "problem" is that I'm generating mutatated form of this protein on some very important residues (Y115E and Y253E), hopping to obtain some change in the conformation in the area of this aminoacids, but at the end they are identical to the wt model (that is identical to the human templates, e.g. PDBID: 2D0T).
I'm also trying to use MD simulation to force a new conformation, but the "wt-like" conformation is very stable in that area, also with mutations.
Creating the mutated model with the wt conformation creates a sort of bias that I can't deal with, until now.
Can you advice me something to deal with this?

Thank you very much, cheers,
Francesco