From: Jeff Comer (jeffcomer_at_gmail.com)
Date: Thu Aug 15 2019 - 17:04:22 CDT
I typically run replica exchange in NVT, after equilibration in NpT.
This might fix your problem, whatever it might be.
How are you accounting for changes in system size between the
replicas? lf you do replica exchange in NpT, in principle, you need to
include a pressure–volume term in the exchange criterion.
Jeffrey Comer, PhD
Institute of Computational Comparative Medicine
Nanotechnology Innovation Center of Kansas State
Kansas State University
Office: P-213 Mosier Hall
On Thu, Aug 15, 2019 at 8:55 AM Giacomo Fiorin <giacomo.fiorin_at_gmail.com> wrote:
> Hi Robyn, that type of error usually occurs early on in the equilibration
> of a system, when the unit cell fluctuations are larger. But especially
> for a smaller system, occasional fluctuations and long-time changes may
> break the patching design that NAMD assign at the start of the run. I've
> seen it happen a few times, regardless of solute scaling.
> Possible workarounds may be increasing margin (unless done so already) or
> use shorter run segments (maybe packed into a long job for the queue
> system) so that the patching subdivision is closer to the most current
> shape of the unit cell.
> On Thu, Aug 15, 2019 at 9:41 AM Gumbart, JC <gumbart_at_physics.gatech.edu>
> > We ran REST2 on a small peptide that frequently crossed boundaries without
> > any issues that I recall ( https://www.ncbi.nlm.nih.gov/pubmed/30888824 )
This archive was generated by hypermail 2.1.6 : Tue Dec 31 2019 - 23:20:54 CST