Dear Amber developers and users,
we are facing unstable simulations while trying to alchemically transform a neutral peptide into a (negatively) charged one, coupled to the transformation of an ion (chloride) into a water molecule (the co-alchemical ion approach from the literature). The simulation shows huge spikes in temperature and in the forces acting on chloride and the coupled water molecule, eventually resulting in out of scale energy components (e.g. EPtot = **************). What is puzzling us is that the problem shows up only when the two transformations are simultaneously performed: independent simulations (one to transform the peptide and another to transform the ion) complete with no issues. We have being playing with the setup with unfortunately no success (different lambda values, different water models, …). I copy below leap, parmed and pmemd input files to run a simulation on a minimal system (in this system I’m transforming a serine into a serine, such that the transformation can also be run without softcore potentials on the peptide, just on the water molecule, showing that in this case the simulation behaves properly). In case, I can provide all files required to run the simulation and also a figure showing the temperature profile and the force on the chloride ion during the simulation.
Any suggestion or advice from the community on how to properly perform the coupled transformation is highly appreciated.
Thank you for your time.
Best regards,
Pietro Vidossich
Laboratory of Molecular Modelling and Drug Discovery
Istituto Italiano di Tecnologia
Genova, Italy
LEAP.IN file:
source leaprc.protein.ff14SB
source leaprc.water.spce
WAT = SPF
loadAmberParams frcmod.spcfw
set default FlexibleWater on
wt = sequence {ACE GLY SER GLY NME}
mut = sequence {ACE GLY SER GLY NME}
model = combine {wt mut Cl- SPF}
setbox model centers {20. 20. 20.}
charge model
saveamberparm model tripep.top tripep.rst
quit
PARMED.IN file:
setOverwrite True
parm tripep.top
loadRestrt tripep.rst
tiMerge :1-5 :6-10 :3 :8
outparm tripep_ti.top tripep_ti.rst
quit
PMEMD.IN file:
co-alchemical ion
&cntrl
imin=0,
ntx=1, irest=0,
ntxo=2, ntpr=1, ntwr=1000, ntwx=1, ntwv=1, ntwf=1, ntwe=1, ioutfm=0,
nstlim=100000, dt=0.001,
ntt=3, temp0=310, ig=-1, tempi=310, gamma_ln=2.0,
ntc=1, nsnb=10,
ntf=1, ntb=1, cut=10.,
clambda=0.00922,
icfe=1, ifsc=1, logdvdl=1,
timask1=':3,7',
timask2=':6,8',
scmask1='.19-22',
scmask2='.43-46,51,52',
/
&ewald
ew_type=0, use_pme=1,
/
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Oct 23 2022 - 01:30:02 PDT
