[gmx-users] segmentation fault with lincs warning

Mon Jul 21 02:56:30 CEST 2008

ANINDITA GAYEN wrote:
> Dear Sir,
> 
>         I have removed now one dmpc molecule from the bilayer, but the problem is the same. I have also tried using a downloaded pdb of chaps from pdb data bank, i.e. 1XU9. Why all the bonds of chaps are rotating more than 30 degree when i have already minimised chaps to a reasonable potential value? Is there any problem in the topology? "Is the topology broken?" i have read that this type of problem may happen if i have problem in my topology file. This can be true as i have built the topology by hand using the toplogy file from prodrg2 server and oplsaa force field bonds, angles, and dihedrals. 
> 

It sounds like your in vacuo minimization was successful, so I doubt that your 
topology is severely broken.  There are plenty of pitfalls in converting a 
GROMOS topology from PRODRG to OPLS-AA, but I recall several discussions on this 
topic a while back, so my comments are the same as those I've already given.

> Can you tell me what will be the rpoblem if i use oplsUA atom descriptions for chaps? Are OPLSUA atom and geometry definintion that are defined in the OPPLSAA itp files in the gromacs ff library totally uncompatible with OPLSAA? 
> 

I know nothing about OPLS-UA, but my knee-jerk reaction to this question is 
"absolutely not."  One cannot represent, say, nonbonded interactions under (for 
example) OPLS-UA, then apply bonded parameters from OPLS-AA, or vice versa, or 
some hybrid of both.

I don't think that messing with the force field representation of CHAPS is going 
to substantially change things.

> 
> If yes, then what is the way? how can i insert chaps in the bilayer?
> 

Alright, so you removed one lipid from your DMPC bilayer, but that does not 
guarantee that all steric clashes are resolved.  Are you still trying to fit 4 
CHAPS into the bilayer, as in your quoted message below, or just one molecule? 
The latter case would probably be the best one.

Also, is this crash occurring during minimization, equilibration, or MD?  Be 
aware also that even if you pass through minimization, not all unfavorable 
clashes will necessarily be resolved by energy minimization.  EM just provides a 
"reasonable" starting structure for the simulation.

If manual removal of a single DMPC does not allow for appropriate insertion of 
CHAPS, try another methodology, such as using genbox or InflatGRO to prepare the 
input structure.

-Justin

> Sincerely
> yours,
> Anindita Gayen
> 
> 
> --- On Mon, 14/7/08, Justin A. Lemkul <jalemkul at vt.edu> wrote:
> 
>> From: Justin A. Lemkul <jalemkul at vt.edu>
>> Subject: Re: [gmx-users] segmentation fault with lincs warning
>> To: aninditagayen at yahoo.co.in, "Discussion list for GROMACS users" <gmx-users at gromacs.org>
>> Date: Monday, 14 July, 2008, 4:40 PM
>> ANINDITA GAYEN wrote:
>>> Dear all,
>>>
>>>
>>> I have an dmpc bilayer equilibrated for 12 ns. I have
>> built a CHAPS molecule in the OPLSAA force field and
>> inserted it in the equilibrated bilayer. There was no
>> problem till the minimisation, but when i started the MD,
>> the job escaped out with a lincs error
>>> Initializing LINear Constraint Solver
>>>   number of constraints is 6572
>>>   average number of constraints coupled to one
>> constraint is 3.3
>>>    Rel. Constraint Deviation:  Max    between atoms   
>>  RMS
>>>        Before LINCS         0.817519   4126   4128  
>> 0.017115
>>>         After LINCS         30.145958  17066  17068  
>> 1.789056
>>>
>>> Step -2, time -0.004 (ps)  LINCS WARNING
>>> relative constraint deviation after LINCS:
>>> max 30.145958 (between atoms 17066 and 17068) rms
>> 1.789056
>>> bonds that rotated more than 30 degrees:
>>>  atom 1 atom 2  angle  previous, current, constraint
>> length
>>>    4128   4129   31.1    0.1571   0.1600      0.1530
>>>   16859  16858   36.1    0.1529   0.1022      0.1529
>>>   16858  16859   36.1    0.1529   0.1022      0.1529
>>>   16897  16858   49.2    0.1523   0.1495      0.1529
>>> [all the atoms and bonds in the error list are for
>> CHAPS]
>>> Do i have the problem in the topology file? I have
>> inserted 4 chaps mole cules without replacing the dmpc
>> molecules, i.e. 4 chaps with 128 dmpc molecules in the
>> bilayer. Is this causing any steric problem?
>> This is a *very* common problem, and a lot of potential
>> solutions are in the 
>> list archive and wiki site.  As such, I won't list them
>> here, but you can 
>> certainly find them within a few minutes (or seconds) of
>> searching.
>>
>> If you have a pre-equilibrated bilayer, and inserted
>> molecules into it without 
>> removing any lipids, you probably have very severe atomic
>> overlap.  But as I 
>> said, search the archive and try the things you find there.
>>
>> -Justin
>>
>>> Please suggest a way to get out of the problem. If you
>> want, i can upload my topology and mdp files in the next
>> mail.
>>> Yours truly,
>>> anindita gayen
>>>
>>>
>>>
>>>
>>>
>>>
>>>       Unlimited freedom, unlimited storage. Get it
>> now, on
>> http://help.yahoo.com/l/in/yahoo/mail/yahoomail/tools/tools-08.html/
>>> _______________________________________________
>>> gmx-users mailing list    gmx-users at gromacs.org
>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>> Please search the archive at
>> http://www.gromacs.org/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/mailing_lists/users.php
>> -- 
>> ========================================
>>
>> Justin A. Lemkul
>> Graduate Research Assistant
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>> ========================================
> 
> 
>       Bring your gang together. Do your thing. Find your favourite Yahoo! group at http://in.promos.yahoo.com/groups/
> 
> 

-- 
========================================

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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