Hi All,
I'm setting up a MD simulation of a solvated ligand with pmemd.cuda in Amber v16. I used "SolvateOct ligand TIP3PBOX 20.0 0.75" to add the water box and I thought the box was large enough (box size 55.7437501 55.7437501 55.7437501; cut=8.0, skinnb=2.0). However I got "the NaN error" at the heating step. Unfortunately there is no error message that I can find. The MD keeps running until the end, but the restart file, output file, and trajectory file all have multiple NaN errors.
Interestingly when I use a larger water box (SolvateOct ligand TIP3PBOX 30.0 0.75) "the NaN error" disappears. I'm a bit confused on how thick the solvate buffer should be. Is this due to the small size of the final box size for pmemd.cuda in my first try even if the 20 A buffer is used?
Here is the end of output file:
===============================================================================
NSTEP = 1000 TIME(PS) = 2.000 TEMP(K) = NaN PRESS = 0.0
Etot = NaN EKtot = NaN EPtot = -45238.6088
BOND = 4.3592 ANGLE = 14.5485 DIHED = 26.1885
1-4 NB = 9.0457 1-4 EEL = -135.6888 VDWAALS = 7634.5270
EELEC = -52795.9780 EHBOND = 0.0000 RESTRAINT = 4.3890
EAMBER (non-restraint) = -45242.9978
------------------------------------------------------------------------------
NMR restraints: Bond = 0.000 Angle = 0.000 Torsion = 0.000
Here is the input file:
Heating up system 50frames 0.1ns cut_8
&cntrl
imin = 0, nstlim = 50000, dt = 0.002,
ntx = 1, irest = 0,
ntb = 1,
ntt = 3, temp0 = 298.0, tempi = 50.0, gamma_ln = 2.0, ig = -1,
ntp = 0,
ntc = 2, ntf = 2,
ioutfm = 1, iwrap = 1,
ntwe = 1000, ntwx = 1000, ntpr = 1000, ntwr = 1000,
cut = 8,
nmropt = 1,
ntr = 1, restraint_wt = 10.00,
restraintmask='!:WAT & !.H=',
/
&ewald
/
&wt
type='TEMP0',
istep1 = 0, istep2 = 8000,
value1 = 50.0, value2 = 298.0
/
&wt type = 'END'
/
Will be happy to share with you all corrupted files off the list if needed.
Jian-xiong Diao
Department of Molecular Biology and Biochemistry,
School of Biological Sciences,
University of California,
Irvine, CA 92697-3900
College of Resources and Environmental Sciences
China Agricultural University,
No. 2 Yuanmingyuan Xi Road Haidian District,
Beijing, P. R. China 100193
Email: d1jx.hotmail.com
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 27 2016 - 16:00:03 PDT
