Dear Modeller users!
I need to refine some flexible regions (mainly long loop and linker regions) of my models using some enhanced sampling engines implemented in modeller like simulated annealing. In this FAQ salilab.org/modeller/6v2/FAQ.html I've found the same method. Could you suggest me some example of the realization of this method in the modeller python script indicating its differences from the method proposed in the advanced tutorial (listed below).
class MyModel(automodel): def special_restraints(self, aln): rsr = self.restraints for ids in (('NH1:161:A', 'O1A:336:B'), ('NH2:161:A', 'O1B:336:B'), ('NE2:186:A', 'O2:336:B')): atoms = [self.atoms[i] for i in ids] rsr.add(forms.upper_bound(group=physical.upper_distance, feature=features.distance(*atoms), mean=3.5, stdev=0.1))
TFH,
James
On 05/26/2014 03:16 AM, James Starlight wrote: > I need to refine some flexible regions (mainly long loop and linker > regions) of my models using some enhanced sampling engines implemented > in modeller like simulated annealing.
OK, just select them with select_loop_atoms and use loopmodel.
> Could you suggest me some example of the realization of this method in > the modeller python script indicating its differences from the method > proposed in the advanced tutorial (listed below).
Sorry, I don't understand this question. All Modeller models are refined by simulated annealing, so there's nothing special about loop modeling in that respect. The main difference between the construction of comparative models and the refinement of loops is that the latter turns on a statistical potential for nonbonded pairs. This is helpful in regions where you have no template structures to provide this information.
Ben Webb, Modeller Caretaker
-
James Starlight -
Modeller Caretaker