Proteins are organic compounds, consisting of amino
acids (residues) bound by peptide bonds into polypeptide
chains. Most of them can fold into their unique functional
structures (native state) without any chaperones. Amino
acids in theoretical protein models are often considered as
beads lying on the sites of a lattice. By extracting
information from the system with different kinds of computer
algorithms, one hopes to predict the folding process.
How and why proteins fold is still not clear although it
has been extensively investigated for more than half a
century. Experimental techniques (NMR, X-ray
crystallography, etc.) as well as computer aided theoretical
work (energy landscape, homology, etc.) have been used to
understand how a protein folds from its amino acid sequence
into a functional structure.
In order to have a more fundamental understanding of
protein structures and their properties, a closer look at a
3D protein shall be taken. Nonnative contact properties for
proteins in a modified model are investigated and compared
with the results from its 2D model analogue. Our computer
program generates all possible conformations (enumeration),
so it enables us to carry out exact calculations for the
nonnative contact density nc(e) as
a function of the energy density e as well as
thermodynamically averaged nonnative contacts
n̄c(T) as a function of
temperature of T. The 3D conformation space can be
generated by setting the distance D as the x-axis,
energy density e as y-axis and entropy density
s as the z-axis. The effect of interaction energies
on the conformation space was also been examined, which can
yield information on how a denatured protein folds to its
native state. These results provide us with a better
understanding of the role that nonnative contacts play in
the protein folding process. Key results are:
1) Compared with the 2D model, nc(e) and n̄c(T) have a
similar behavior in 3D for all three models (described in
the text) investigated here but with more continuous shape,
because more energy levels are available in the 3D
model.
2) Odd or even residue numbers will affect the n̄c(T) for a
protein with given number of residues in 2D and 3D.
3) Changing the interaction energy parameters will aect
greatly the conformation space of the protein.