Problem with Protein-DNA docking

Dear All,
I am working on protein DNA interactions. After the docking of the protein DNA using HADDOCK 2.4 the resultant docked PDB doesn’t contain the same active sites as it was given during the file submission and the DNA which is supposed to be in a particular site and in particular orientation according to a PDB identical with more than 50%, changed in the opposite direction.
Please suggest ways to overcome this problem.

There is no magical solution to that. If you want to obtain exactly the same complex, then define very specific distance restraints. Did you also look at different clusters?

What do you mean by this?

It seems like you already know how the protein should interact, you can define more specific restraints to ensure the correct orientation, have a look at this topic about atom-specific restraint definitions.