AMBER Archive (2000)

Subject: Gibbs and PME simulation with DNA

From: David Bosch (david_at_caos.uab.es)
Date: Mon Jul 31 2000 - 08:20:02 CDT


Dear AMBER users,

We are intending to run a PMF simulation in a 6-mer of DNA just varying two
dihedral angles of one strand and using PME on a R12000 platform. We use
AMBER5. We are aware of bufix.16 about GIBBS and PME (published on the AMBER
home page) so we fixed the ewald.f and recompiled. Even though our calculations
do not show any kind of vlimit error and the PMF energies obtained seem
reasonable the geometries are completely wrong with flipped bases and clashes.

Any help form anybody that successfully ran GIBBS simulations of double
stranded DNA with PME will be wellcome. Is there any other bugfix that we have
to take into account to circumvent this problem?.

Thanks in advance for your help!!

David Bosch, Ph.D student

Lab. Medicina Computacional
U. Bioestadistica, Facultat de Medicina
Universitat Autonoma de Barcelona
08193 Bellaterra
CATALUNYA

--