[gmx-users] Re: Re: quality of cluster strucutres (David van der Spoel)

[Mauricio Sica]

> Mauricio Sica wrote:
> > Dear all
> >
> > I´ve obtained one cluster structure from a protein simulation. Then, as
I
> > wanted to use this structure as a static model, I checked it with
WHAT_IF
> > and the result was that as a model is a little poor (a list of bad
angles,
> > bad distances, problems with planarity and so on as you can imagine). I
> > decided to minimize it .So, I performed simulated annealing
(300>360>300)
> > PRMD (only C_alfas in posre.itp) and then minimization. But the model
> > continues didn't improve. I know this would be expected for a
"particular"
> > conformation at a " particular" moment during the simulation.
> > Have any one of you any suggestion?
> >
> you don't say HOW you got the structure
>
> g_cluster -method gromos suggests a "real" structure from the trajectory
> as best representative for each cluster.
>
> >
I used gromos as well as linkage methods and the results were the same