[gmx-users] A question on LINCS error after significant simulation time around 400ps.
feenstra at chem.vu.nl
Mon Feb 9 08:53:01 CET 2004
Hwankyu Lee wrote:
> Dear, gmx-users,
> I'm running peptide(SP-B(1-25),including aromatics) simulations with monolayers.
> I put peptide, which is used with "-dummy hydrogens" option, into monolayers
> with different angles with respect to interface between water and monolayers,
> and then I got LINCS error from one of them around 400ps, although other
> simulations were running nicely until 500ps.
> although only difference among them is different angle between peptide and
> interface of monolayers and water.
First off, 'only' a different angle is still quite a big difference in atomic
coordinates, relative orientations of peptide vs. lipid tails, etc.
As for Lincs warnings, they happen sometimes for no apparent reason. My
workaround is always to try a re-start from just before the errors, and see
if they reoccur. If they do, there is a problem in your system, e.g. bad
contacts or some other instability.
Of course, an exact restart will give the same errors, therefore I introduce
slight round-off errors in the coordinates at the 6th decimal place, using:
#> trjconv -f traj.trr -s topol.tpr -o frame300.gro -ndec 6 -dump 300
#> trjconv -f frame300.gro -o frame300.trr
#> tpbconv -f frame300.trr -s topol.tpr -o restart.tpr
'trjconv -ndec 6' writes a .gro file with 6 decimal places (in stead of the
usual 3), which is nearly single precision. '300' is, obviously, the time
from which you want to restart. Then it is converted to .trr so that tpbconv
will accept it, and finally tpbconv will generate a new .tpr file for your
| | |
| _ _ ___,| K. Anton Feenstra |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|( | )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610 |
| | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/ |
| | "If You See Me Getting High, Knock Me Down" |
| | (Red Hot Chili Peppers) |
More information about the gromacs.org_gmx-users