[gmx-users] system blowing up

Sat Jul 28 12:43:49 CEST 2012

 I ran for the fourth time and got the error again. Now, I think the random output of such a system is not reliable!

Sincerely,
Shima

----- Original Message -----
From: Shima Arasteh <shima_arasteh2001 at yahoo.com>
To: Discussion list for GROMACS users <gmx-users at gromacs.org>
Cc: 
Sent: Saturday, July 28, 2012 3:06 PM
Subject: Re: [gmx-users] system blowing up

Thanks for your reply.

I'm following the protocol of Diagnosing an unstable system. Corresponding the first step I changed th nstxout to 1 for 3 times.
The first and second time I got error the same as before ( some interactions seem to be assigned multiple times ) and it the warning as below was appeared:

Warning: 1-4 interaction between 5 and 18 at distance 2.017 which is larger than the 1-4 table size 2.000 nm

For the third time, I did'nt get any error and the output files were generated without any warning or error. 

Is it acceptable to go on with these outputs? Why such a thing happened? Is the system minimized sufficiently after 2 times?

Thanks 

Sincerely,
Shima

________________________________
From: Mark Abraham <Mark.Abraham at anu.edu.au>
To: Discussion list for GROMACS users <gmx-users at gromacs.org> 
Sent: Saturday, July 28, 2012 6:12 AM
Subject: Re: [gmx-users] adding a new atom type

On 28/07/2012 4:34 AM, Shima Arasteh wrote:
>
> As I explained here before, I get an error ( some interactions seem to be assigned multiple times).

And it's been suggested several times that your system may be 
http://www.gromacs.org/Documentation/Terminology/Blowing_Up, and 
diagnostic and management strategies are described there. Since you 
haven't said whether it was immediately upon simulation start (so that 
the topology might be doubtful, and probably only if you are the first 
to find some new bug) or later on (when the topology is not directly to 
blame for blowing up). Messing around duplicating atom types to try to 
avoid the situation is presupposing the form of the solution to a 
problem that you have not yet identified.

>   I suspected the top file, so checked it  and visualized the output of em.mdp; the output seems to be correct and I don't see any bonds behaves strangely.

OK, but the "bonds" you see there do not relate to the ones in your 
topology. 
http://www.gromacs.org/About_Gromacs/Related_Software/Visualization_Software. 
So all you've seen is that nothing's gone wrong enough so far.

> As I studied in mailing lists it and you also mentioned , this may happen because of the insufficient minimization. I'm wondering how a good and sufficient minimization  is provided?

Often whatever comes out of a single-precision EM run is fine. You have 
not presented evidence that your minimization is insufficient, and there 
are many other possible causes for blowing up. In the absence of a 
description of your system, your .mdp file and how you've modified your 
forcefield and/or topology, it's impossible to get effective help.

Mark

>
>
>
> Sincerely,
> Shima
>
>
> ----- Original Message -----
> From: Justin Lemkul <jalemkul at vt.edu>
> To: Shima Arasteh <shima_arasteh2001 at yahoo.com>; Discussion list for GROMACS users <gmx-users at gromacs.org>
> Cc:
> Sent: Friday, July 27, 2012 10:51 PM
> Subject: Re: [gmx-users] adding a new atom type
>
>
>
> On 7/27/12 1:08 PM, Shima Arasteh wrote:
>>
>> Which parameter is supposed to be changed to get the sufficient minimized system? Would anyone guide me  please?
>>
> This question is very vague.  In principle, any system can be minimized but the
> definition of "sufficient" depends on your goals.
>
> -Justin
>
>> Thanks in advance
>>
>> Cheers,
>> Shima
>>
>>
>> ----- Original Message -----
>> From: Justin Lemkul <jalemkul at vt.edu>
>> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>> Cc:
>> Sent: Wednesday, July 25, 2012 4:37 PM
>> Subject: Re: [gmx-users] adding a new atom type
>>
>>
>>
>> On 7/25/12 7:58 AM, Shima Arasteh wrote:
>>>
>>>       Some time ago I defined a new residue and after running grompp I got an error as there are some interactions assigned multiple times. Then I decided to duplicate the atoms involved in that residue to avoid this error.
>> That error usually indicates your system is crashing and the DD algorithm is failing.  That means either the coordinates are insufficiently minimized, the topology is incorrect, or the .mdp settings are leading to instability.
>>
>>> Next I renamed the atomtypes in atometypes.atp as below:
>>>       ZC    12.01100 ;    carbonyl C, peptide backbone
>>> ZCT1    12.01100 ;    aliphatic sp3 C for CH
>>> ZCT3    12.01100 ;    aliphatic sp3 C for CH3
>>> ZO    15.99900 ;    carbonyl oxygen
>>> ZHC    1.00800 ;    N-ter H
>>> ZNH1    14.00700 ;    peptide nitrogen
>>> ZH    1.00800 ;    polar H
>>> ZHB    1.00800 ;    backbone H
>>> ZHA    1.00800 ;    nonpolar H
>>>
>>> Then I wanted to modify the .itp file, got in to the trouble; I couldn't find the C atomtype in nonbonded.itp file. Actually I don't know what to do now!
>>>
>> Start over with a fresh set of force field files that you haven't experimented with and troubleshoot the stability of your system.  If you Google or search the mailing list archive for this error, tons of useful results show up.
>>
>> -Justin
>>
>> -- ========================================
>>
>> Justin A. Lemkul, Ph.D.
>> Research Scientist
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>> ========================================
>> -- gmx-users mailing list    gmx-users at gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> * Only plain text messages are allowed!
>> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>> * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-request at gromacs.org.
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>

-- 
gmx-users mailing list    gmx-users at gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Only plain text messages are allowed!
* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-request at gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists 