AMBER Archive (2000)

Subject: improper density in constant pressure md

From: Scott A. Showalter (sashowal_at_wanda.wustl.edu)
Date: Thu Oct 12 2000 - 13:18:51 CDT

Hi,

We have recently begun using the parallel version of Amber on an IBM with
AIX unix. (We previously ran single processor jobs on an SGI.) I have
successfuly run constant pressure MD in the past using input similar to
the one shown below, but for some reason all of the test runs I have
performed using the parallel code converge to a density of about
0.85. Here is the input script:

Constant Pressure MD to equilibrate Water
 &cntrl
        imin=0, irest=0, ntx=1,
        ntf=2, ntc=2, ntb = 2,
        ntp=1, npscal=1, cut=10,
        nstlim=25000, dt=0.002,
        ibelly=1, taup=1.0,
        ntpr=100, ntwr=100, ntwx=100,
 &end
Group information for all non-protein atoms
ATOM 1650 25283
END
END

I know that using belly and PME at the same time is technically incorrect,
but it has never produced problems for these types of equilibration runs
before. Is there a problem with my input file which somebody could point
out?

Thank you for your input

Scott