One correction:
I was able to get normal outputs with “logdvdl = 0, idecomp = 1”, but not vice versa.
->
I’m able to get normal outputs with “logdvdl = 1, idecomp = 0”, but not “logdvdl = 0, idecomp = 1”
From: He, Amy via AMBER <amber.ambermd.org>
Date: Friday, January 27, 2023 at 6:14 PM
To: AMBER Mailing List <amber.ambermd.org>
Subject: [AMBER] idecomp = 1 for TI simulations: Energies not printed / cannot be calculated
Dear Amber Experts,
I am using Amber 20 and sander.MPI for dual-topology TI simulations with soft core potential. I encountered a problem with “idecomp = 1”. I seemed to be able to “start” the simulation, but my mdout file is incomplete and does not have any output data for the simulation. I don’t think I’m actually able to “run” the simulation.
I figured that the issue was caused by “idecomp = 1”. I was able to get normal outputs with “logdvdl = 0, idecomp = 1”, but not vice versa. With “idecomp = 1”, I did not get mden or a trajectory either. I’m sharing a sample of my mdout, a sample of md input and my group file.
Is idecomp trying do a per-residue (water included) energy decomposition? I’m simulating a protein with explicit waters, so if idecomp was originally built for PB or GB simulations then that could be a super costly option for explicit solvent simulations..
Normally, the DV/VL contributions of the soft core atoms are printed in the mdout. However, I wish to see the DV/VL contribution of a few other residues (like “the background electrostatics” or “conformational rearrangement” components). Is that possible?
Am I using a depreciated option/protocol for TI simulations? :”(( Any comments/suggestions would be greatly appreciated.
Many Thanks,
Amy
The input files are as attached. Below is the plain text version of them:
========== sander command ==========
mpirun -np 40 $AMBERHOME/bin/sander.MPI -ng 2 -groupfile prd.group -inf prd.info
(I shouldn’t do -inf prd.info!)
========== mdout ==========
(the last a few lines are)
| MPI Timing options:
| profile_mpi = 0
Noshake mask :202,446; matches 55 atoms
Removing shake constraints from H HID 446 -- N HID 446
(a few similar lines….)
Removing shake constraints from H5 REI 202 -- C3 REI 202
LOADING THE DECOMP ATOMS AS GROUPS
========== mdin ==========
step 1
&cntrl
imin = 0, irest = 1, ntx = 5, dt = 0.002,
nstlim = 500000,
ntwe = 1000, ntwx = 1000, ntpr = 1000, ntwr = 5000,
ntb = 2, ntp = 1, pres0 = 1.0, taup = 1.0, barostat = 1,
ig = -1, ntt = 3, gamma_ln = 2.0, temp0 = 310.0, tempi = 310.0,
ntc = 2, ntf = 1,
noshakemask = ':202,446',
ioutfm = 1, iwrap = 1,
icfe = 1, ifsc = 1, scalpha = 0.5, scbeta = 12.0,
clambda = 0.1,
scmask = ':REI.N4,P1,H28',
logdvdl = 0, idecomp = 1,
/
&ewald
/
========== group file ==========
-O -i prd0.in -p v0_combine.prmtop -c press0.rst7 -ref v0_combine.inpcrd -inf prd0.info -o prd0.out -r prd0.rst7 -e prd0.en -x prd0.nc
-O -i prd1.in -p v1_combine.prmtop -c press1.rst7 -ref v1_combine.inpcrd -inf prd1.info -o prd1.out -r prd1.rst7 -e prd1.en -x prd1.nc
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jan 27 2023 - 15:30:03 PST
