[Pictured below is an average structure taken over 1 nanosecond from
a solvated particle mesh Ewald simulation of a decamer poly(A)-poly(T)
duplex. The average structure was created using carnal
to RMS fit and coordinate average all DNA atoms from the trajectory taken
at 1 picosecond intervals.]
DNA:DNA, polyA-polyT
The purpose of this tutorial is to demonstrate how to set up a
standard decamer poly(A)-poly(T) duplex DNA model structure using the
tools provided with AMBER 4.1 and more generally provide information
on setting up and running AMBER.
In this discussion, we first figure out how to generate a starting
structure and then use this structure to build up the input files
necessary for sander. The basic files necessary to run
sander are:
- prmtop: a description of the molecular topology and force
field parameters
- inpcrd (or a restrt from a previous run):
a description of the coordinates and optionally the velocities
and current box dimensions
- mdin: the sander input file which is a series of
namelists and control variables; old timers can still resort
to formatted input if you like that kind of thing...
After we've built up the prmtop and inpcrd files,
for in vacuo and solvated systems, we will actually run
sander to perform minimization and molecular dynamics and
eventually get to the point where we can create the picture shown
above. This tutorial will include a description of one possible way
to "equilibrate" the system and then move into a brief discussion of
performing "production" molecular dynamics runs. After initial
equilibration, all of the simulations will be run using the
particle mesh Ewald (PME) method
[
J. Chem. Phys. 103, 8577-8593. (1995)
]
within AMBER 4.1. Some trajectory analysis
issues with the current version of the PME code will
also be discussed.
thomas <cheatham@cgl.ucsf.edu>>
Last modified: Thu Aug 29 12:27:09 1996