[gmx-developers] strange forces in MD of protein-lipid system

[Rainer Böckmann]

I just found out that the lipid being extracted from the membrane is one 
that is split over two procs when using 8 procs. We'll investigate this 
further.
Best,
Rainer


David van der Spoel wrote:

>On Fri, 22 Sep 2006, Rainer Böckmann wrote:
>
>  
>
>>Dear Developers,
>>
>>we found strange forces acting on one lipid in a mixed protein-lipid 
>>system (~24k) running on 8 processors (3.3.1, PME), resulting in this 
>>lipid leaving the membrane on the timescale of 10-100ps (stable 
>>simulation). This force acts only on atoms in a specific position of the 
>>pdb-file, i.e. exchanging two lipids will result in a different lipid 
>>leaving the membrane. Also, restart on a different number of processors 
>>avoids this artefact. If required we can support the respective tpr-files.
>>    
>>
>
>are you sure this is the case? is it not pbc?
>
>otherwise, is it not due to an error in the topology? there is recent 
>bugzilla on strange effects due to the topology.
>
>in addition, do you have the pme_order bug fixed, for pme_order != 4?
>
>if you think this is a bug, submit a bugzilla.
>
>  
>
>>Best,
>>rainer
>>
>>
>>    
>>
>
>  
>


-- 
__________________________________________________________
Dr. Rainer Böckmann
Theoretical & Computational Membrane Biology
Center for Bioinformatics Saar
Universität des Saarlandes
Gebäude C7.1 (17.1), EG
D-66041 Saarbrücken, Germany
Phone: ++49 +681 302-64169  FAX: ++49 +681 302-64180
E-Mail: rainer at bioinformatik.uni-saarland.de
http://www.bioinf.uni-sb.de/RB/
___________________________________________________________