[gmx-users] segmentation fault in position restrained step

[Justin A. Lemkul]

Moeed wrote:
> Dear Justin,
> 
> 1- Could you please check if I have grouped atoms properly? Lastly, I 
> could generate the tpr file for PR step. Since I am getting segmentation 
> fault in the next step I though maybe there is sth wring with charge 
> groups..

I don't know how your labeling is set up.  If your CH3 and CH2 groups are all 
your charge groups, then you should be fine.  In fact, this is what grompp 
suggested in the note.  You can get a sense of what might be appropriate by 
looking at your force field's .rtp file and looking at the functional groups.

> 
> However, I have a funda,mental question. I am to compute interaction 
> parameters for ternary system of hexane/polyethylene/ethylene. Sofar I 
> have only hexane as solvent, later polyethylene and ethylene will be 
> added. My questions is for this apolar system, do I need to worry about 
> electrostatic interactions between atoms? I mean could I skip buildig 
> charge groups if I am interested in cacluation of interaction parameters 
> BETWEEN hexane, polyethylene and ethylene?

Your atoms have partial charges, do they not?  Then you certainly need to 
consider proper electrostatics treatment.

> 
> [ atoms ]
> ;   nr       type  resnr residue  atom   cgnr     charge       mass  
> typeB    chargeB      massB
>      1   opls_157      1   HEX      C1      1      -0.18     12.011   ; 
> qtot -0.18
>      2   opls_158      1   HEX      C2      2      -0.12     12.011   ; 
> qtot -0.3
>      3   opls_158      1   HEX      C3      3      -0.12     12.011   ; 
> qtot -0.42
>      4   opls_158      1   HEX      C4      4      -0.12     12.011   ; 
> qtot -0.54
>      5   opls_158      1   HEX      C5      5      -0.12     12.011   ; 
> qtot -0.66
>      6   opls_157      1   HEX      C6      6      -0.18     12.011   ; 
> qtot -0.84
>      7   opls_140      1   HEX      H1      1       0.06      1.008   ; 
> qtot -0.78
>      8   opls_140      1   HEX      H2      1       0.06      1.008   ; 
> qtot -0.72
>      9   opls_140      1   HEX      H3      1       0.06      1.008   ; 
> qtot -0.66
>     10   opls_140      1   HEX      H4      2       0.06      1.008   ; 
> qtot -0.6
>     11   opls_140      1   HEX      H5      2       0.06      1.008   ; 
> qtot -0.54
>     12   opls_140      1   HEX      H6      3       0.06      1.008   ; 
> qtot -0.48
>     13   opls_140      1   HEX      H7      3       0.06      1.008   ; 
> qtot -0.42
>     14   opls_140      1   HEX      H8      4       0.06      1.008   ; 
> qtot -0.36
>     15   opls_140      1   HEX      H9      4       0.06      1.008   ; 
> qtot -0.3
>     16   opls_140      1   HEX     H10      5       0.06      1.008   ; 
> qtot -0.24
>     17   opls_140      1   HEX     H11      5       0.06      1.008   ; 
> qtot -0.18
>     18   opls_140      1   HEX     H12      6       0.06      1.008   ; 
> qtot -0.12
>     19   opls_140      1   HEX     H13      6       0.06      1.008   ; 
> qtot -0.06
>     20   opls_140      1   HEX     H14      6       0.06      1.008   ; 
> qtot 0
> 
> 2- I tried to run the position restrained simulation. VERSION 4.0.7 
> with  "mdrun -s Hexane_pr.tpr -o Hexane_pr.tpr -c Hexane_b4md -v >& 
> output.mdrun_pr". After a few seconds I get segmentation fault error. I 
> did a thorough search on mailing list and found a smilar situation where 
> Mr. MArck Abraham introduced  
> http://oldwiki.gromacs.org/index.php/blowing_=up but link is not 
> working. Do I need to change the constraint algorithm? Could you please 
> tell me why I am getting segmentation error? Does it have to do with 
> kinetic energy?
> 

Did you do energy minimization?  Usually instabilities like this arise because 
either the system is energetically unstable due to atomic clashes, or something 
about the underlying model physics is broken.  You haven't mentioned how you 
built your system or if you energy minimized it, so I assume that you simply 
haven't resolved the clashes in your system.

<snip>

> define              =  -DPOSRES

If your system is just a box of hexane, restraining anything doesn't make sense 
to me.  One usually employs position restraints on a solute (like a protein) to 
relax the solvent (usually water) around it.  If you're trying to equilibrate a 
hexane system, you're just wasting your time by restraining any or all of the 
molecules.

-Justin

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

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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