Marco Neves wrote: > I have two pdb files of the same protein in complex with different ligands. > PDB1 has all residues. PDB2 misses a small loop. > Is it possible to model the missing loop of PDB2 based on the > corresponding residues of PDB1 while keeping all other residues in the > exact same positions? > Which functions should be used?

You can certainly build a model of your target using both PDB1 and PDB2 as templates - just align PDB1 with your target in that loop region.

As for using loop modeling, that isn't really applicable here because loop modeling is designed for cases where you have no template - it won't use any information from PDB1. If you want to use the standard loop modeling protocol though, and just use the conformation from PDB1 as your starting loop, you could override the loopmodel.build_ini_loop() method. You'd have to first do a local superposition of PDB1 onto your model so that the loop is in the correct orientation though, so I am skeptical whether this would buy you anything.

Ben Webb, Modeller Caretaker