Dear all,
I am having some trouble with TI calculation and I'm hoping to get some
pointers on how to solve this problem.
What I am trying to do is to basically mutate a sugar to a hydroxyl group.
The issue arises in the van der Waals step using softcore potentials (step
2). I can minimize, heat, equilibrate and run production calculations for
the charging steps (steps1 & 3). My input files look like this:
v0:
&cntrl
imin = 0, ntx = 1, irest = 0,
ntpr = 1 , ntwr = 1000, ntwx = 1,
ntf = 1, ntc = 2, ntr =1, nmropt = 1,
ntb = 1, cut = 9.0, ig = -1,
nstlim = 50000, dt = 0.001,
ntt = 3, gamma_ln = 5,
icfe=1, clambda = 0.1,
ifsc=1,
crgmask=':4.C1
,H1,O5,C5,H5,C6,H61,H62,O6,H6O,C4,H4,O4,H4O,C3,H3,O3,H3O,C2,H2,O2,H2O',
scmask=':4.C1
,H1,O5,C5,H5,C6,H61,H62,O6,H6O,C4,H4,O4,H4O,C3,H3,O3,H3O,C2,H2,O2,H2O',
&end
&wt
TYPE='TEMP0', istep1=0, istep2=25000, value1=5.0, value2=300.0,
/
&wt
TYPE='TEMP0', istep1=25001, istep2=50000, value1=300.0, value2=300.0,
/
&wt
TYPE='END',
/
Keep nucleoside fixed with weak restraints
1.0
RES 1 7
END
END
v1:
temperature equilibration
&cntrl
imin = 0, ntx = 1, irest = 0,
ntpr = 1 , ntwr = 1000, ntwx = 1,
ntf = 1, ntc = 2, ntr = 1, nmropt = 1,
ntb = 1, cut = 9.0, ig = -1,
nstlim = 50000, dt = 0.001,
ntt = 3, gamma_ln = 5,
icfe=1, clambda = 0.1,
ifsc=1,
crgmask=':4.HOD',
scmask=':4.HOD,
&end
&wt
TYPE='TEMP0', istep1=0, istep2=25000, value1=5.0, value2=300.0,
/
&wt
TYPE='TEMP0', istep1=25001, istep2=50000, value1=300.0, value2=300.0,
/
&wt
TYPE='END',
/
Keep nucleoside fixed with weak restraints
1.0
RES 1 7
END
END
I keep getting vlimit errors. I have read several messages on the list about
this, tried different clambda and also removed all restraints but the
error persists. Minimization runs without any issues. Also when I use dt =
0.0005 it seems to run (I have done about 500ps, although the temperature of
v1 system remains ~ 600K which I have seen other people mention in the list
archives). Using a restart from the dt = 0.0005 calculation does not solve
the problem.
When I examine the structures of the calculations that crash with this
vlimit/SHAKE errors I find that strangely it sometimes does not occur due to
the residue that is been changed.
I created my initial structures from a glycosylated peptide that has been
equilibrated by normal MD. To make the second (mutated) structure with the
hydroxyl instead of the sugar I manually deleted the atoms corresponding to
the sugar from a pdb file (except the glycosidic oxygen) and reloaded the
pdb in xleap to add the hydrogen atom. There is a net charge on the peptide
during the TI transformations but the only way I can see to have a neutral
system is to mutate whole residues which has the same vlimit errors when I
try.
I am using GLYCAM force field for the sugar and standard amber SCEE/SCNB
but I would think if this was the issue it might be more important in steps
1 and 3.
Any ideas on how to tackle this is very welcome.
Thanks,
Emmanuel
-------------------------------------------------------------------------------------------------------------------------------------------------
v0 output before crashing
-------------------------------------------------------------------------------------------------------------------------------------------------
NSTEP = 697 TIME(PS) = 498.894 TEMP(K) = 319.49 PRESS =
839.1
Etot = -21886.5090 EKtot = 5835.0153 EPtot =
-27721.5243
BOND = 314.7526 ANGLE = 361.4444 DIHED =
73.3269
1-4 NB = 43.7350 1-4 EEL = 243.7492 VDWAALS =
4405.6716
EELEC = -33164.2039 EHBOND = 0.0000 RESTRAINT =
0.0000
DV/DL = 7179.5437
EKCMT = 3160.1045 VIRIAL = 1458.3109 VOLUME =
93927.6476
Density =
0.9755
Ewald error estimate: 0.2712E-04
------------------------------------------------------------------------------
Softcore part of the system: 22 atoms, TEMP(K) =
495.20
SC_BOND= 15.5740 SC_ANGLE= 32.4067 SC_DIHED =
5.7785
SC_14NB= 5.9273 SC_14EEL= 0.0000 SC_EKIN =
27.0614
SC_VDW = -1.4267 SC_EEL = 0.0000 SC_DERIV= -7.7901
------------------------------------------------------------------------------
vlimit exceeded for step 697; vmax = 37.2853
Coordinate resetting (SHAKE) cannot be accomplished,
deviation is too large
NITER, NIT, LL, I and J are : 0 5 28 32 33
Note: This is usually a symptom of some deeper
problem with the energetics of the system.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 26 2011 - 11:30:04 PDT
