Hi folks,
I'm setting up TI calculations with Cando using AMBER.
I'm having a problem where about 1/3rd of the TI calculations are failing
with the error message below:
ERROR: Calculation halted. Periodic box dimensions have changed too much
from their initial values.
(more below)
Could someone point me at a good protocol for carrying out TI calculations
that avoids this? Or confirm that it's ok to add a constant pressure CPU
pmemd run between the heating and TI steps.
I am following the protocol described in the AMBER/FEP tutorial (
http://ambermd.org/tutorials/advanced/tutorial9/#overview)
The steps that I'm using and copied from the benzene/phenol in lysozyme
example are:
(1) Create the complex and ligand topology/crd files.
(2) Minimize the energy of each system.
(3) Heat each system to 300K
(4) Pressurize each system to 1atm.
(5a) For the vdw calculation, for each lambda window, heat each system to
300K.
(6a) (Here it blows up) For the vdw calculation, for each lambda window,
run a TI calculation.
(5b) For decharge/recharge stages strip the ligands from the system and
construct new systems that decharge and recharge the appropriate ligands.
(6b) For the decharge/recharge stages, for each lambda window heat each
system
(7b) (Here it blows up). For the decharge/recharge stages, for each lambda
window, run a TI calculation.
The problem is that a lot of the TI calculations are failing with the error
message below:
ERROR: Calculation halted. Periodic box dimensions have changed too much
from their initial values.
Your system density has likely changed by a large amount, probably from
starting the simulation from a structure a long way from equilibrium.
[Although this error can also occur if the simulation has blown up for
some reason]
The GPU code does not automatically reorganize grid cells and thus you
will need to restart the calculation from the previous restart file.
This will generate new grid cells and allow the calculation to continue.
It may be necessary to repeat this restarting multiple times if your
system
is a long way from an equilibrated density.
Alternatively you can run with the CPU code until the density has
converged
and then switch back to the GPU code.
They are blowing up in the TI step and I'm not sure why, the minimization,
heating, pressurization, and then heating steps for the TI windows appear
fine.
I'm looking for ideas on how to debug the calculation.
Here is an example of what I'm working with:
[image: image.png]
[image: image.png]
This is one of the FEP calculations - the wire-frame is the common parts of
the molecule and the ball-and-stick are the differences (scmask1 and
scmask2 respectively). The receptor is 3L9H - a kinesin.
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Tue Aug 06 2019 - 19:00:02 PDT
