The advance of computers has made rendering of molecules
in three dimensions relatively easy and common place. There
are several ways one can display a model of any particular
molecule and I will briefly discuss the way we are displaying
them in both the GUI (client, newly installed nodes and
release version 2.0) and on the webpage.
1. Stick model: Here the molecule is effectively displayed
by sticks that represent bonds between atoms. The atoms
themselves are not shown. The thickness of the stick can
be changed to taste. This type of model is particularly
useful to chemists (scientist that make these molecules)
since important structural features are easily identified.
Often molecules (candidates) are shows this way. The atoms
are colored in a consistent manner:
C - carbon (black or grey)
H - hydrogen (white)
O - oxygen (red)
N - nitrogen (blue)
An example is given below:
Figure A: Example of a stick model.
2. Ball and stick model: This is similar to the stick
model with the atom themselves being represented as balls
of larger diameter than the sticks (bonds). This type of
model does not yield more information than the stick model.
Chemists tend to favor the stick model for its simplicity.
Atoms add clutter to the picture but are esthetically pleasing
to look at. This type of model is regularly used by scientist
who build plastic models of molecules to obtain a better
idea of the 3-D nature of the molecule they are dealing
with.
Figure B: Example of a ball and stick
model.
3. CPK - space filling model: CPK are the initials of
the scientist that first used this type of model (Corey,
Pauling, Koltun). Here the atoms are represented as overlapping
spheres which diameters corresponding to their Van der Waals
radius. This model gives a better sense of how crowded the
space is around molecules. This type of model is not very
useful to chemist since much of the structure is not visible
from any perspective. There are however cases where it is
important to understand what the spatial requirements are
for a molecule to fit into a pocket or fit through a hole
in a membrane.
Figure C: Example of a space filling
model. (CPK)
4. Ribbon: Proteins are very large molecules made up of
repetitive subunits called amino acid residues and are referred
to as hetero-polymers. The sequence of amino acid residues
strung together in a long polymer is often referred to as
its primary structure. This long string forms several different
types of secondary structures such as alpha helices (springs),
beta pleated sheets and beta turns. The highest level of
organization in a protein is referred to as its tertiary
structure and refers to the arrangement of various secondary
structural elements such as alpha helices, beta pleated
sheets and beta turns. The ribbon model of a protein shows
the three dimensional arrangement secondary structural elements
and is a flat ribbon like representation of the backbone
of the string of amino acid residues. This type of model
gives less specific detail about the arrangement of individual
atoms in the structure but displays important information
with regards to the tertiary structure of the protein. Often
a combination of stick and ribbon models are used to display
proteins. For comparison a stick model, a ribbon and stick
model and a ribbon model of a protein are shown below:
Figure D: Stick model of human Rhinovirus
main protease
Figure E: Ribbon and stick model of
human Rhinovirus main protease
Figure F: Ribbon model of human Rhinovirus
main protease
The different secondary structural motives can be identified
by their coloration:
Red - alpha helices
Turquoise - beta pleated sheets
Green - beta turns
Surface model: In particular the solvent accessible surface
(ie. The surface of the molecule as experienced by a proto-typical
solvent molecule) is useful to probe the surface of a protein
for pockets and crevices as potential binding pockets. Superimposed
on this surface is a visualization of partial atomic charges
which portraits the electronic requirements for binding
in that region. The sign and magnitude of the partial charges
are indicated by color and intensity respectively. The brighter
the color the higher the charge.
Blue - Negative charge
Red - Positive charge
White - Neutral
This model looks similar to the CPK representation but communicates
significantly more valuable information.
Figure G: Solvent accessible surface
of human Rhinovirus main protease colored according to
partial charges on solvent accessible atoms.
The surface model is most informative for binding sites
and we will use this type of representation for the binding
sites under investigation on all the targets. The rest of
the protein (target) will be displayed as a ribbon model.
This yields the most information and is visually most pleasing.
An example is given below:
Figure H: Ribbon and surface structure
of human Rhinovirus main protease.
To this "pocket" one can then add the inhibitor ( = drug
candidate: molecules that inhibit a function of a protein
(target) are often potential drugs). An Example is given
below:
Figure I: Ribbon and Surface structure
of human Rhinovirus main protease with an inhibitor (AG7051)
bound in the active site.
this page
last reviewed October 1, 2004
