[gmx-users] Fumbling in the dark with dihedral pulling

[Justin Lemkul]

On 4/5/20 5:05 PM, Marko Petrovic wrote:
> Hello.
>
> I've been trying to do coordinate pulling of dihedral angles of an alanine dipeptide in vacuum and used the umbrella sampling tutorial as a starting point. I'm trying to generate initial configurations for later runs, but the mdrun ends in a segmentation fault with error messages like:
>
> Step 15707, time 31.414 (ps)  LINCS WARNING
> relative constraint deviation after LINCS:
> rms 0.025328, max 0.098100 (between atoms 11 and 13)
> bonds that rotated more than 30 degrees:
>   atom 1 atom 2  angle  previous, current, constraint length
>        7      8   90.4    0.1006   0.1054      0.0997
>       11     13   90.2    0.1110   0.1220      0.1111
>
> Step 15708, time 31.416 (ps)  LINCS WARNING
> relative constraint deviation after LINCS:
> rms 52572448.000000, max 238546496.000000 (between atoms 7 and 8)
> bonds that rotated more than 30 degrees:
>   atom 1 atom 2  angle  previous, current, constraint length
>        7      8  109.3    0.1054 23783084.0000      0.0997
>        9     11  119.5    0.1543 1995948.6250      0.1538
>       11     12  162.6    0.1127 1995948.6250      0.1111
>       11     13  115.6    0.1220 1995948.5000      0.1111
>       11     14  112.7    0.1124 1995948.6250      0.1111
> Wrote pdb files with previous and current coordinates
> Segmentation fault: 11
>
> Perhaps the seg fault is not related to the warnings at all but as I am not certain what they mean and do think they are enough to cause some problems anyway I would like to understand them in general and also know what columns previous and current are saying.

They refer to bond lengths, what the length was at the previous step and 
what it is now, compared to what the reference value should be. Your 
values indicate catastrophic distortion of the structure.

> As the original tutorial was done for a water solution and mine is conducted in vacuum I am guessing my modifications to the .mdp files are not completely updated so any tips on that woul also be helpful.

There's no way to provide any useful advice without you posting your 
.mdp file. Note that the umbrella sampling tutorial has basically no 
relevance to anything you're doing because enforced dihedral rotation 
didn't exist when I wrote it :)

-Justin

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Office: 301 Fralin Hall
Lab: 303 Engel Hall

Virginia Tech Department of Biochemistry
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

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