Dear Mingyuan,
At first glance your input files look OK.
Because you are not using a periodic system it may be best to set qmcut to a large value, for instance 999.0 which will effectively include all MM atoms in the QM/MM Hamiltonian. But this is likely not reason for the problems that you encounter.
You are using a time step of 1fs, which may be too large. Try 0.5 fs.
Another reason for your problems may be large differences in the water equilibrium structure between the QM method (DFTB3) and the water force field (you must be using some flexible water model, since you are not using SHAKE constraints). You can try to get this under control by stronger coupling to the thermostat (increase gamma_ln). It may be possible that it would require a very large transition region or will not work at all if differences in the potentials are too large.
All the best,
Andy
—
Dr. Andreas W. Goetz
Assistant Research Scientist
San Diego Supercomputer Center
Tel: +1-858-822-4771
Email: agoetz.sdsc.edu
Web: www.awgoetz.de
> On Dec 18, 2017, at 8:39 AM, Mingyuan_xu <ming_yxu.163.com> wrote:
>
> Dear Amber members:
> I have meet several questions about Amber adaptive QM/MM MD. I am doing a QM/MM umbrella sampling to scan the potential energy surface and the QM theory for A region is DFTB3. I found out that several water molecules fluctuate very fast and the DFTB3 calculations could not achieve convergent result after 50 time steps,then the sander bombed。Before each QM/MM calculations, I have run several picoseconds MM MD to make sure the structure normal. My input file md_equil.in is shown as follows:
>
>
> 50 ps equilibration for Zn and COO
> &cntrl
> imin = 0, ntx = 1, irest = 0,
> ntpr = 1, ntwr = 1, ntwx = 1,
> ntf = 1, ntc = 1, cut = 999.0,
> ntb = 0, nstlim = 10000, dt = 0.001,
> tempi=300.0, temp0 = 300.0, ntt = 6,
> gamma_ln = 1.0,igb=0,
> nmropt = 1, ioutfm=0,
> ivcap=0,
> fcap=20.0,
> ntr=1,restraintmask='.5 ',restraint_wt=300.0,
> ifqnt=1
> &end
> &qmmm
> qmmask=':1,2',# the Zinc ion and a COO-
> qmcharge=1
> spin=1
> vsolv=2
> adjust_q=0
> qmcut=8.0
> qm_theory='dftb3'
> &end
> &vsolv
> nearest_qm_solvent=10
> qm_center_atom_id=5 # atom index of CD atom in COO-
> &end
> &adqmmm
> n_partition =10
> verbosity=1
> print_qm_coords=1
> /
> &wt
> type='END',
> &end
> DISANG=disZNC.in
>
>
> And my prmtop file and initial coordinate file are listed in accessory. To solve this problem, I have test different values of n_partition parameter, but the results are almost the same。
> Could you help me? It's really my honor to receive any letter from you，thank you 。
> Best wishes for you!
> Mingyuan
> 2017.12.18<COO_PMF_20.parm7><disZNC.in><md_heat.rst>_______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Dec 22 2017 - 18:00:02 PST
