Hello,
I am using pmemd.cuda_DPFP (Amber 20) to do minimization for my system
with both positional and NMR distance restraints.
The NMR distance restraints are used to pull one residue close to the
substrate during the minimization gradually (force constant gradually
increased in five steps of minimizations).
And the positional restraints are applied to heavy atoms except for the
substrate and the region where the residue locates (20 residues).
For two almost identical systems, the input files are the same. But one
works for all five steps, pmemd.cuda_DPFP can distinguish both
positional and NMR distance restraints, as the output shows:
#
RESTRAINT = 705.1138 (positional restraints)
NMR restraints: Bond = 595.904 Angle = 0.000 Torsion =
0.000 (NMR distance restraints)
#
While the other one failed to distinguish positional and NMR distance
restraints:
#
RESTRAINT = 1175.6790
NMR restraints: Bond = 1175.679 Angle = 0.000 Torsion =
0.000
#
For the failed system, the first step of minimization worked fine with
force constant of 1 kcal/mol A2,
but it failed to distinguish at the second step of minimization with a
force constant of 10 kcal/mol A2.
And it seems that positional restraints are on the region where the
residues locates (20 residues), though I did not turn it on,
leading to failure of pulling the residue close to the substrate.
Here is my input files:
#
&cntrl
imin = 1, ntmin = 2,
maxcyc = 25000, ncyc = 15000,
ntpr = 100, ntwe = 100,
ntb = 1, lj1264 = 0 , nmropt = 1 ,
ntr = 1, restraint_wt = 100,
restraintmask=' ( :1-189 | :209-263 | :267-455 | :475-529 ) & !.H= ',
/
&ewald
/
&wt type='DUMPFREQ', istep1=100, /
&wt type='END', /
DISANG=dists2.RST
DUMPAVE=dists2.dat
LISTIN=POUT
LISTOUT=POUT
#
I tried pmemd.cuda (single version), it worked, but the minimized
structure was a little twisted, some bonds are at length around 2 angstrom).
I could not figure out the issue. Any comments or suggestions? Thanks a
lot!
All the best,
Qinghua
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Mar 01 2021 - 17:30:01 PST
