Dear Modeller users,

I am using Modeller6.2. To test the quality of the model, I tried to build a model for a protein (300 aa)using its crystal structure as the template. With MD_level = refine_1 or 3, the backone RMSD is 0.25-0.3, which is great. However, in both refinement conditions, the RMSD of the heavy atoms of the backbone plus chainchains is close to 3.0 angstroms. By visual inspection of the superimposed structures, the sidechains of the model indeed do not overlap with its crystal structure so well.

My questions: Did you observe similar result? How do you refine the sidechains? MD simulation of the sidechains with backbone restrained (in explicit water solution) is in my mind. I would appreciate your suggestion and experience.

Regards,

Jain Hui Wu Lady Davis Institute McGill Universty