Dear Amber community,
I am trying to do TI in Amber14 with pmemd but I've noticed that at high lamda values (somewhere > 0.95) the compounds become distorted, I get 'vlimit exceeded for step' warnings and the simulation crashes. This happens both during heating (which can't finish for the high lambda values) and during production runs as well, which, obviously, also can't finish. I'm trying to do a one-step transformation using softcore potentials.
Heating:
mpirun -np 2 pmemd.MPI -i heat.in -c start.rst7 -ref start.rst7 -p ligands.parm7 -O -o heat.out -inf heat.info -e heat.en - heat.rst7 -x heat.nc -l heat.log
The input file:
heating
&cntrl
imin = 0, nstlim = 10000, irest = 0, ntx = 1, dt = 0.001,
ntt = 1, temp0 = 300.0, tempi = 50.0, tautp = 1.0,
ntc = 1, ntf = 1,
ntb = 1,
ioutfm = 1, iwrap = 1,
ntwe = 1000, ntwx = 1000, ntpr = 1000, ntwr = 5000,
icfe = 1, clambda = 1.0, scalpha = 0.5, scbeta = 12.0,
logdvdl = 0,
timask1 = ':1', timask2 = ':2',
ifsc=1,
scmask1=':1XJ.C4,C5,C14,C15,C17,C24,C25,C26,C28,C29,C30,C37,C38,C39,C40,C41,C42,C47,N49,O58,S63,Cl6,F59,F60,F61,H4,H5,H14,H15,H49,H281,H282,H291,H292,H301,H302,H371,H372,H381,H382,H401,H402,H403,H411,H412,H413,H421,H422',
scmask2=':1X3.N1,N93,O1,C81,C82,C83,C4,C5,C86,C87,C88,C89,C90,H35,H36,H81,H82,H83,H4,H5,H86,H87,H88,H89',
crgmask='',
/
&ewald
/
&wt
type='TEMP0',
istep1 = 0, istep2 = 8000,
value1 = 50.0, value2 = 300.0
/
&wt type = 'END'
/
Production TI command:
mpirun -np 2 pmemd.MPI -i ti.in -c heat.rst7 -p ligands.parm7 -O -o ti001.out -inf ti001.info -e ti001.en -r ti001.rst -x ti001.nc -l ti001.log
The input file:
TI simulation
&cntrl
imin = 0, nstlim = 100000, irest = 0, ntx = 1, dt = 0.001,
ntt = 3, temp0 = 300.0, gamma_ln = 2.0, ig = -1,
ntc = 1, ntf = 1,
ntb = 2,
ntp = 1, pres0 = 1.0, taup = 2.0,
ioutfm = 1, iwrap = 1,
ntwe = 100, ntwx = 100, ntpr = 100, ntwr = 100,
icfe = 1, clambda = 1.0, scalpha = 0.5, scbeta = 12.0,
logdvdl = 1,
ifmbar = 1, bar_intervall = 1000, bar_l_min = 0.0, bar_l_max = 1.0,
bar_l_incr = 0.1,
timask1 = ':1', timask2 = ':2',
ifsc=1,
scmask1=':1XJ.C4,C5,C14,C15,C17,C24,C25,C26,C28,C29,C30,C37,C38,C39,C40,C41,C42,C47,N49,O58,S63,Cl6,F59,F60,F61,H4,H5,H14,H15,H49,H281,H282,H291,H292,H301,H302,H371,H372,H381,H382,H401,H402,H403,H411,H412,H413,H421,H422',
scmask2=':1X3.N1,N93,O1,C81,C82,C83,C4,C5,C86,C87,C88,C89,C90,H35,H36,H81,H82,H83,H4,H5,H86,H87,H88,H89',
crgmask=''
/
&ewald
/
All simulations are stable and produce sensible results with the corresponding sander setup and a lambda interval from 0.1 to 0.9. I would like to sample the end states, however, so I'm trying to get the simulation to run with pmemd.
Any help would be immensely appreciated.
Sincerely,
Stefan
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jan 27 2017 - 03:30:03 PST
