1. Introduction
AMMP is a modern full-featured molecular mechanics, dynamics
and modelling program. It can manipulate both small molecules and macromolecules
including proteins, nucleic acids and other polymers. In addition to standard
features, like numerically stable molecular dynamics, fast multipole method for
including all atoms in the calculation of long range potentials and robust
structural optimizers, it has a flexible choice of potentials and a simple yet
powerful ability to manipulate molecules and analyze individual energy terms.
One major advantage over many other programs is that it is easy to introduce
non-standard polymer linkages, unusual ligands or non-standard residues. Adding
missing hydrogen atoms and completing partial structures, which are difficult
for many programs, are straightforward in AMMP. It is written in C and has been
ported to many different computers. The source code is available under the GNU "copyleft"
so that it is easy to follow exactly what the program is doing. Another major
advantage over commercial software is that AMMP can be easily included in
another program to supply molecular mechanics function.
The main features implemented in AMMP are:
Fast long range electrostatics and non-bonded terms. AMMP runs without cutoffs in times comparable to or better than an 8 Angstrom cutoff in other programs.
Stable, numerically accurate, molecular dynamics. AMMP is stable with difficult problems such as parallel stranded DNA. (shown with its Na counterions below).
Flexible choice of potential functions. In addition to standard formulations, AMMP has non-point charges, explicit Debye screening, accurate polarization models, non-harmonic bond and angle formulations, and coupled bond-angle terms available.
Calculates partial charges for new molecules.
Embedding and homotopy methods for rapid model building.
Neural network polymer modeller (Kohonen neural network).
Self-contained, and can be embedded in other programs.
VEGA ZZ interface.