David Gell wrote:
i am trying to build models for a single missing loop on a known xtal structure and i have a couple of questions 1) can modeller introduce hydrogen bonds in loop modelling (is it true that all-hydrogen models are not fully implemented for loop modelling)? 2) is it true that modeller does not include hydrophobic packing factors? i currently have models with fully exposed hydrophobics that have similar 'modeller objective function' to those with more buried hydrophobic sidechains. i'm a modeller newbie, and above is based on my understanding (i.e. likely flawed) of the mail archives and docs. perhaps i should be introducing some 'artificial' distance restraints to mimic the effects of the above interactions?
You are exactly correct on all points. The loop modeling potential acts only on heavy atoms - there are no explicit hydrogen bonds. There is also no explicit consideration of hydrophobic packing - this is implicit in distance restraints inherited from the template, and is thus missing for loop regions. You could, however, use the dope_loopmodel class rather than loopmodel, which includes GB/SA implicit solvation, to remedy this somewhat.
As you say, you can certainly add your own distance restraints if you are aware of any specific interactions.
Ben Webb, Modeller Caretaker -- modeller-care@salilab.org http://www.salilab.org/modeller/ Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage