Dear Amber users,
I am running two parallel Amber calculations for (1) an RNA and (2)
the same RNA with a small molecule ligand. The run for the free RNA
works really well, especially when I set saltcon=0.2. Previously I had
saltcon=0.1 and both the free RNA and the complex worked reasonably
well but with a higher RMSD for the ensemble. When I try to switch to
saltcon=0.2 for the complex I get a lot of vlimit exceeded message and
the energies shoot up very rapidly to reach "NaN" and no output
coordinates obviously. Right now I am testing higher vlimit settings
but I know that this is not a solution since even if the run did
finish the structures may not be relevant. I am including my input
file below and would be grateful for any suggestions. Is it something
to do with moving the small molecule into place which is harder at
higher saltcon?
Thanks,
Michael.
&cntrl
nstlim=50000, pencut=-0.001, nmropt=1,
ntpr=500, ntt=1, ntwx=200,
cut=15.0, ntb=0, vlimit=50,
igb = 1, iscale=10, imin=0,
dt=0.0002, saltcon=0.2, gbsa=1,
/
&wt type='TEMP0', istep1=0,istep2=12500,value1=500.,
value2=500., /
&wt type='TEMP0', istep1=12501, istep2=45000, value1=500.0,
value2=100.0, /
&wt type='TEMP0', istep1=45001, istep2=50000, value1=0.0,
value2=0.0, /
&wt type='TAUTP', istep1=0,istep2=12500,value1=0.4,
value2=0.4, /
&wt type='TAUTP', istep1=12501,istep2=45000,value1=4.0,
value2=4.0, /
&wt type='TAUTP', istep1=45001,istep2=47500,value1=1.0,
value2=1.0, /
&wt type='TAUTP', istep1=47501,istep2=50000,value1=0.1,
value2=0.05, /
&wt type='REST', istep1=0,istep2=7500,value1=0.1,
value2=1.0, /
&wt type='REST', istep1=7501,istep2=50000,value1=1.0,
value2=1.0, /
&wt type='IMPROP', istep1=0,istep2=50000,value1=1,
value2=250, /
&wt type='BOND', istep1=0,istep2=50000,value1=1,
value2=4, /
&wt type='ANGLE', istep1=0,istep2=50000,value1=1,
value2=4, /
&wt type='TORSION', istep1=0,istep2=50000,value1=1,
value2=4, /
&wt type='END' /
LISTOUT=POUT
DISANG=./REFINE.RST
DIPOLE=./1.anneal/DIPOLE.RST
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Feb 25 2010 - 08:30:02 PST
