[gmx-users] Constraints in the villin system

[Mario Fernández Pendás]

Thank you very much Justin.

2015-06-01 18:10 GMT+02:00 Justin Lemkul <jalemkul at vt.edu>:

>
>
> On 6/1/15 11:55 AM, Mario Fernández Pendás wrote:
>
>> Dear all,
>>
>> I am interested in simulations with the villin system using the constraint
>> algorithm SHAKE.
>>  From the physical point of view, does it make any sense to perform a
>> simulation only with the hydrogen atoms constrained?
>>
>>
> Bonds involving hydrogen, not the atoms themselves, yes.  That's fairly
> standard procedure for most simulations.
>

I have seen in the literature many cases where all atoms are constrained so
that is why I considered that maybe my set-up could be meaningless.

>
>  Otherwise, when I try to constraint all the atoms using PBC and domain
>> decomposition, the LINCS algorithm is suggested and I am still interested
>> in using SHAKE.
>>
>>
> You can certainly try it, but SHAKE is not as stable as LINCS and has
> problems satisfying that many simultaneous constraints.
>

I am interested in SHAKE's reversibility feature and that is why I am using
it.

Thank you.

Cheers,
Mario