Dear AMBER community,
I encounter the below error message when trying to perform a minimization. The input file is at the very end of the message.
NSTEP ENERGY RMS GMAX NAME NUMBER
1 NaN NaN 2.1274E+77 N 8047
BOND = 1627.2953 ANGLE = 5004.2201 DIHED = 6967.7007
VDWAALS = NaN EEL = NaN HBOND = 0.0000
1-4 VDW = 1947.9588 1-4 EEL = 30599.1262 RESTRAINT = 0.3922
EAMBER = NaN
DV/DL = NaN
NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.392
===============================================================================
Softcore part of the system: 37 atoms, TEMP(K) = 0.00
SC_Etot= 0.0000 SC_EKtot= 0.0000 SC_EPtot = 138.6008
SC_BOND= 9.0733 SC_ANGLE= 21.7073 SC_DIHED = 63.9746
SC_14NB= 6.6511 SC_14EEL= 79.3772 SC_VDW = -1.4378
SC_EEL = -40.7448
SC_RES_DIST= 0.0000 SC_RES_ANG= 0.0000 SC_RES_TORS= 0.0000
SC_EEL_DER= NaN SC_VDW_DER= -Infinity SC_DERIV = NaN
------------------------------------------------------------------------------
| TI region 2
NSTEP ENERGY RMS GMAX NAME NUMBER
1 NaN NaN 2.1274E+77 N 8047
BOND = 1627.2953 ANGLE = 5004.2201 DIHED = 6967.7007
VDWAALS = NaN EEL = NaN HBOND = 0.0000
1-4 VDW = 1947.9588 1-4 EEL = 30599.1262 RESTRAINT = 0.3922
EAMBER = NaN
DV/DL = NaN
NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.392
===============================================================================
Softcore part of the system: 37 atoms, TEMP(K) = 0.00
SC_Etot= 0.0000 SC_EKtot= 0.0000 SC_EPtot = 118.4809
SC_BOND= 5.3882 SC_ANGLE= 17.0508 SC_DIHED = 40.9427
SC_14NB= 2.7221 SC_14EEL= 8.6995 SC_VDW = -1.3387
SC_EEL = 45.0164
SC_RES_DIST= 0.0000 SC_RES_ANG= 0.0000 SC_RES_TORS= 0.0000
SC_EEL_DER= NaN SC_VDW_DER= -Infinity SC_DERIV = NaN
------------------------------------------------------------------------------
Coordinate resetting (shake) was not accomplished
within 3000 iterations
*** Especially for minimization, try ntc=1 (no shake)
The offending atom (N 8047) is the backbone N of the to-be mutated residue and it is in the softcore region). If I follow the suggestion of entirely turning of SHAKE, the offending atom changes to CG 8054 (still in the softcore region) and the minimization also crashes on the first step. (I know both cases crash just after the first step because I was printing every step (ntpr=1).
Visually I don’t see any bad sterics, and clambda was set to 0. Values of clambda of 0.5 or 1.0 also produce crashes on the first step, but the offending atoms are different and not so many energy terms are NaN.
Curiously, out of 17 calculations generated from different snapshots of the system from an initial-state trajectory, only this calculation, and one other, crashed with the above error. The offending atom in the other crashed calculation was also N 8047.
I want to understand why these two calculations in particular crashed. Are there any suggestions of what I can try to figure this out? My aim is to take a large number of snapshots from trajectories of the initial and final states, re-build the topology for TI simulations, minimize the system, and then proceed with fast-growth morphing simulations. I don’t want a pile of snapshots that couldn’t be used for some unidentified reason.
Thanks for your help!
Input file:
&cntrl
ig=-1, ! Random seed (=-1, get a number from current date and ti
imin=1, ! Perform an energy minimization
ntmin=2, ! Method of minimization
ncyc=1000, ! Number of steepest descent cycles
maxcyc=1000, ! Maximum number of minim
ntb=1, ! Boundaries (=1, constant volume)
igb=0, ! GB model (=0, explicit solvent)
cut=10.0, ! Cutoff
ntc=2, ! SHAKE constraints (=2, hydrogen bond lengths constraine
ntf=1, ! Force evaluation (=2, hydrogen bond interactions omitte
ntr=0, ! no positional restraints
nmropt=1, ! To vary target temperature below
icfe=1, ! Turn on free energy calcs
ifsc=1, ! Use softcore potentials
scalpha=0.5, ! alpha parmaeter in vdW decoupling
scbeta=12, ! Å^2; scaling parmaeter for softcore electorstatics
gti_cut=1, ! Normal non-bonded cuttoff not applied among softcore
gti_chg_keep=1, ! Do not neutralize charges on softcore atoms
gti_lam_sch=0, ! Do not use lambda scheduling
gti_ele_sc=0, ! Do not use smoothstep function for electrostatics
gti_vdw_sc=0, ! Do not use smoothstep function for vdW interactions
gti_add_sc=1, ! Scale 1-4 non-bonded terms between common and softcore
gti_bat_sc=0, ! Do not scale bonded terms between common and softcore b
gti_output=0, ! Default output
gti_cpu_output=1, ! Combine and output lambda-derivative terms for all TI r
clambda=0.0, ! Value! Value of perturbation parameter
dynlmb=0, ! Increment clambda every ntave steps in single morphing
ntave=0, ! Number of steps for each clambda in single morphing sim
klambda=1, ! Exponent in perturbed potential function
tishake=1, ! Normal use of SHAKE
noshakemask='.3630-3644,8047-8068,4389-4410,8813-8827',
timask1='.3630-3644,8047-8068',
timask2='.4389-4410,8813-8827',
scmask1='.3630-3644,8047-8068',
scmask2='.4389-4410,8813-8827',
crgmask= , ! Set to zerggo charges on mask
ti_vdw_mask= , ! Set to zero vdW for mask
logdvdl=0, ! Print dv/dl info
ntwr=1, ! Restart file written every ntwr steps
ntwx=1, ! Trajectory file written every ntwx steps
ntpr=1, ! The mdout and mdinfo files written every ntpr steps
ntxo=2, ! Write NetCDF coordinate file
ioutfm=1, ! Trajectory file format (=1, Binary NetCDF)
iwrap=0, ! Translate water molecules into the original simulation
/
&wt type='DUMPFREQ', istep1=100 /
&wt type='END' /
DISANG=InterSubunitRestraints.rst
DUMPAVE=InterSubunitRestraints_TImin.dat
Best,
Matthew
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 31 2022 - 18:30:03 PDT
