[gmx-developers] [RFC] Test for full cycle of md simulation of small peptide

[Justin Lemkul]

On Mon, Dec 16, 2013 at 5:34 AM, Alexey Shvetsov <alexxy at omrb.pnpi.spb.ru>wrote:

> Hi!
>
> May be it will be good to have small test that emulates full cycle of md
> system preparation and run (just for few dozens of steps for each mdrun).
> As small peptide we can use e.g. speptide, cpeptide or may be even
> trp-cage. This will allow to catch bugs in preparation utils (e.g. genbox,
> genion)
>
>
That could be good, but how robust would such a test be?  Energies could
vary quite a bit based on ion placement (which by default is random) and I
have even seen cases where genbox adds a slightly different number of
waters to identical solutes in identical boxes, so the topology may not
even be consistent.  This is fairly rare, but I do get people panicking
when doing my tutorials if they somehow get one fewer water molecule in
their box ;)

-Justin

-- 

==========================================

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441


==========================================
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20131216/860496b1/attachment.html>