28
Bioorganic Chemistry
that determine the changes in energy due
to deviation from the ideal bond length, an-
gle, and so on. The parameters of a force
Feld are determined in multiple ways, in-
cluding experimental techniques (e.g. ideal
bond lengths from crystal structures, bond
stretch force constants from spectroscopy)
and high-level quantum mechanical cal-
culations where no experimental values
are available. There are many force Felds
available, including MM±±94, CHARM,
AMBER, OPLS, MM2/MM3. Each is con-
structed and parameterized using different
approaches and therefore each gives dif-
ferent results for different systems. Again,
before overinterpreting the results given
by a force Feld in analyzing a new system,
it is important to assess the accuracy of the
given force Feld in predicting experimen-
tally known values in a related system.
Once a molecule is ‘‘drawn up’’ using a
modeling package such as Sybyl, Spartan,
or Hyper-Chem, it may have nonoptimal
bond angles, distances, and so on. Energy
minimization can then be used to ‘‘relax’’
themoleculeintoalocalenergyminimum.
Energy minimization takes a multidimen-
sional derivative of the energy with respect
to a given atomic motion and determines
the forces on a given atom. These forces
sumtopu
shth
ea
tom
sin
toth
e
i
rlo
c
a
l
lowest energy position. This is a local and
not a global energy minimum, because it is
the nature of energy minimization that the
molecule is always going ‘‘down-hill’’ ener-
getically, and therefore it will be unable to
traverse energetic barriers that may allow
access to a lower energy conformation.
Why would a bioorganic chemist use
molecular modeling? Often, it will be to
understand what a molecule looks like,
to determine, for example, how it may
interact with a target macromolecule. ±or
this purpose, it is of greater use to identify
the global energy minimum structure,
that is, the conformation of the molecule
that will have the greatest population.
This problem can be approached using
a variety of tools, including systematic
conformational searching (SCS), Monte
Carlo searching, and molecular dynamics.
Since a majority of the movement in
a molecule is due to rotations around
bonds, SCS looks for the global energy
minimum
by
generating
all
possible
conformers by incrementally advancing
the dihedral angles of the molecule. This
approach will deFnitely explore all possible
conformations (within the resolution of
the dihedral angle increment), but the
number of conformations necessary to
examine increases exponentially with each
additional rotatable bond. Eventually, this
can become computationally unwieldy.
To address this problem, Monte Carlo
searches also rotate bonds in an attempt to
traverse high-energy barriers but do so in a
random manner that is biased away from
conformers that are too high in energy.
There is a trade-off at play here: The Monte
Carlo search will not be as thorough as a
systematic search (except at the highest
number of cycles) but will avoid the very
highest energy regions of confomer space,
which the SCS will be forced to examine.
An additional method of searching for
the global energy minimum of a molecule
is simulated annealing, a tool of molecular
dynamics. In molecular dynamics, the
same
force
Feld
used
to
model
the
static energy of a given conFguration of
atoms is used to determine its dynamic
nature. Momentum is imparted to each
atom (the amount being dependent on
the ‘‘temperature’’ at which the dynamic
simulation
is
run)
and
the
resulting
motions of atoms simulated using the
force Feld. In simulated annealing, a
high temperature is simulated (i.e. a large
amount of momentum applied to the
previous page 702 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online next page 704 Encyclopedia of Molecular Cell Biology and Molecular Medicine read online Home Toggle text on/off