Re: [AMBER] presumed problem with SHAKEH: disappears on passing from CUDA version to normal Sander

From: Scott Le Grand <varelse2005.gmail.com>
Date: Thu, 6 Sep 2012 09:07:27 -0700

Send me the input files? I'm betting you have some some of invalid SHAKEH
network that the CPU code does not correctly detect.

On Thu, Sep 6, 2012 at 2:08 AM, Emanuele Monza <emanuele.monza.bsc.es>wrote:

> Hello,
>
> I am trying to run a MD of hemoglobin. After preparing all necessary input
> with Xleap I minimize the sistem with Sander (four different minimizations)
> and everything it is fine. Then I have four stages of equilibration (100 ps
> each). These are ran with CUDA version. The simulation dies giving the
> following error message:
>
> *Hydrogen atom 219 appears to have multiple bonds to atoms 220 and 221
> which is illegal for SHAKEH.
> Exiting due to the presence of inconsistent SHAKEH hydrogen clusters.*
>
> There is a problem: Atom 219 is NOT a hydrogen, is a carbon... So I try to
> run normal Sander (no CUDA acceleration) and everything it is fine!
> I am using Amber 11 and preparing the system with Amber Tools 12.
> Do you have any idea of which could be the problem? Thanks in advance.
>
> Best regards,
> Emanuele
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 06 2012 - 09:30:04 PDT
Custom Search