[gmx-users] Enforced rotation errors

[Nash, Anthony]

Dear All,

I hope you can help. I am using 'flex' enforced rotation to rotate a cylindrical protein along the surface of a globular protein. Unfortunately my system is experiencing what I can only assume is an IO problem:

DD  step 949999 load imb.: force  2.9%  pme mesh/force 0.677

           Step           Time         Lambda
         950000     1900.00000        0.00000

   Energies (kJ/mol)
          Angle    Proper Dih.  Improper Dih.          LJ-14     Coulomb-14
    1.57412e+04    1.99606e+04    9.25263e+02    7.98071e+03    8.61345e+04
        LJ (SR)  Disper. corr.   Coulomb (SR)   Coul. recip. Position Rest.
    9.59944e+05   -8.46602e+04   -6.08542e+06    6.95493e+04    1.22242e+03
   COM Pull En.      Potential    Kinetic En.   Total Energy    Temperature
    3.91316e+02   -5.00823e+06    8.56761e+05   -4.15147e+06    3.10369e+02
 Pres. DC (bar) Pressure (bar)   Constr. rmsd
   -4.21942e+02    3.27704e+00    2.75540e-05


-------------------------------------------------------
Program mdrun_mpi, VERSION 5.0.5
Source code file: /home/columbus/chem_software/GROMACS/2015/gromacs-5.0.5/src/gromacs/pulling/pull_rotation.c, line: 2502

Fatal error:
Enforced rotation: No reference data for first slab (n=-13), unable to proceed.
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------


I am using gromacs 5.0.5. The .mdp settings are:

; ENFORCED ROTATION
; Enforced rotation: No or Yes
rotation                 = Yes
; Output frequency for angle, torque and rotation potential energy for the whole group
rot-nstrout              = 1
; Output frequency for per-slab data (angles, torques and slab centers)
rot-nstsout              = 10
; Number of rotation groups
rot-ngroups              = 1
; Rotation group name
rot-group0               = Collagen_CA
; Rotation potential. Can be iso, iso-pf, pm, pm-pf, rm, rm-pf, rm2, rm2-pf, flex, flex-t, flex2, flex2-t
rot-type0                = flex
; Use mass-weighting of the rotation group positions
rot-massw0               = yes
; Rotation vector, will get normalized
rot-vec0 = 0.691585601358247 -0.6995947474399045 0.17965674311989613

; Pivot point for the potentials iso, pm, rm, and rm2 [nm]
;rot-pivot0               = 4.31852e+00  1.73201e+00  1.89800e+00

; Rotation rate [degree/ps] and force constant [kJ/(mol*nm^2)]
rot-rate0                = 0.021
rot-k0                   = 600.0 ; testing these
; Slab distance for flexible axis rotation [nm]
rot-slab-dist0           = 1
; Minimum value of Gaussian function for the force to be evaluated (for flex* potentials)
rot-min-gauss0           = 0.001
; Value of additive constant epsilon' [nm^2] for rm2* and flex2* potentials
rot-eps0                 = 0.0001
; Fitting method to determine angle of rotation group (rmsd, norm, or potential)
rot-fit-method0          = norm
; For fit type 'potential', nr. of angles around the reference for which the pot. is evaluated
rot-potfit-nsteps0       = 21
; For fit type 'potential', distance in degrees between two consecutive angles
rot-potfit-step0         = 0.25

The following files are being generated:

flex_1.trr
flex_1.edr
flex_1.log
flex_1.xvg
#flex_1.log.1#
#flex_1.log.2#
#flex_1.log.3#
flex_1_out   <--output from the cluster

Any help is greatly appreciated.
Kind regards
Anthony

Dr Anthony Nash
Department of Chemistry
University College London