Re: AW: Turning off and back on fixed atoms

From: Brian Radak (brian.radak.accts_at_gmail.com)
Date: Sat Oct 29 2016 - 12:04:57 CDT

No, it does not. But we could potentially put this on the list of new
features. Just permitting the fixed atom list to be reallocated
on-the-fly might be a relatively minor modification if it is implemented
like I expect (no promises on that though!).

What I still don't understand is how such a usage pattern would make
sense - what special handling of those degrees of freedom would need to
be implemented? My assumption is that the reason fixed atoms are
currently static is that the forces and velocities on them are just
arbitrarily zeroed (or never integrated, same thing). This is, however,
not a rigorous constraint because there is no formal modification of the
Hamiltonian. Is this a deal breaker?

Brian

On 10/28/2016 04:40 AM, Norman Geist wrote:
>
> Are you sure that namd-2.12 allows to re-read the fixedatomsfile?
>
> *Von:* owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] *Im
> Auftrag von *Radak, Brian K
> *Gesendet:* Donnerstag, 27. Oktober 2016 19:05
> *An:* namd-l_at_ks.uiuc.edu; Norman Geist <norman.geist_at_uni-greifswald.de>
> *Betreff:* RE: namd-l: Turning off and back on fixed atoms
>
> Actually getting NAMD to permit toggling of features is actually not
> too tough - the question is whether or not the actual forces, etc. are
> still being correctly computed as a result.
>
> I'm not familiar with Tiger2A, but the obvious concern is in how one
> properly handles a fluctuating number of degrees of freedom. Is there
> a proposed solution to this problem? What is the desired behavior for
> things like the virial?
>
> My guess is that you will have to intermittently read, write, and
> reload a PDB file specifying the fixed selection and then reallocate
> the structure from the PSF (using reinitatoms) - this should be
> possible in 2.12, but it is still a beta feature for sure.
>
> Brian Radak
> Postdoctoral Appointee
> Leadership Computing Facility
> Argonne National Laboratory
>
> 9700 South Cass Avenue, Bldg. 240
> Argonne, IL 60439-4854
> (630) 252-8643
> brian.radak_at_anl.gov <mailto:brian.radak_at_anl.gov>
>
> ------------------------------------------------------------------------
>
> *From:*owner-namd-l_at_ks.uiuc.edu
> <mailto:owner-namd-l_at_ks.uiuc.edu>[owner-namd-l_at_ks.uiuc.edu] on behalf
> of Norman Geist [norman.geist_at_uni-greifswald.de]
> *Sent:* Thursday, October 27, 2016 5:14 AM
> *To:* namd-l_at_ks.uiuc.edu <mailto:namd-l_at_ks.uiuc.edu>
> *Subject:* namd-l: Turning off and back on fixed atoms
>
> Hi,
>
> Id like to implement the Tiger2A REMD via the TCL interface, using
> the common API replicaSend/Recv etc. This would require to be able to
> turn on and off fixed atoms and should also allow to change which
> atoms are fixed, between run commands.
>
> I know this is not possible so far, but I guess people would
> appreciate this feature.
>
> Best wishes
>
> Norman Geist
>

-- 
Brian Radak
Postdoctoral Scholar
Gordon Center for Integrative Science, W323A
Department of Biochemistry & Molecular Biology
University of Chicago
929 E. 57th St.
Chicago, IL 60637-1454
Tel: 773/834-2812
email: radak_at_uchicago.edu

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:46 CST