B RIEFINGS IN BIOINF ORMATICS . VOL 10. NO 3. 247^258
Advance Access publication March 30, 2009
doi:10.1093/bib/bbp013
2D molecular graphics: a flattened
world of chemistry and biology
Peng Zhou and Zhicai Shang
Submitted: 31st December 2008; Received (in revised form) : 17th February 2009
Abstract
Molecular graphics provides an intuitive way for representation, modeling and analysis of complex chemical and
biological systems. It is now widely used in the theoretical chemistry, structural biology, molecular modeling and
drug design communities. Traditional molecular graphics techniques mainly dedicate to showing molecular architectures at three-dimensional (3D) level. However, in some occasions the two-dimensional (2D) representation of
molecular configurations, profiles, behaviors and interactions may be more readily acceptable for audiences, especially when we need to describe abstract information in a straightforward way or to present numerous data in
schematic diagrams. In recent years, 2D representation methods/tools have been developed rapidly for various
purposes, ranging from the aesthetic depiction of atomic arrangement for small organic molecules to schematic
layout of complicated nonbonding network across the biomolecular binding interfaces, and have received considerable interest in the fields of chemistry, biology and medicine. In this article we first propose the term of 2D
molecular graphics to cover the spectrum of 2D representing chemical and biological systems, we also give a
comprehensive review on the methods, tools and applications of 2D molecular graphics.
Keywords: 2D molecular graphics; schematic diagram; chemical and biological system; bioinformatics
INTRODUCTION
Graphic language, symbolic language and written
language represent the human communication ways
at different abstract levels, in which the graphic
language presents information in an intuitive way
and, apparently, is the most understandable manner
for audiences—‘a picture is worth a thousand words’
[1]. In the fields of chemistry, biology and medicine,
molecular graphics is widely used for study and
dissemination of molecular structures and functions.
It is assumed that microscopic world could be
visualized on computer screen using a dozen of
metaphors, such as the use of spheres to represent
atoms or ribbons to represent protein chains, because
we are generating synthetic images of objects that
are far smaller than the wavelength of light [2].
Traditionally, the subject of ‘molecular graphics’
is always associated with the stereo representation
of molecular architectures in three-dimensional (3D)
space by using various models such as CPK, BALL &
STICK, and SURFACE, etc. [3]. In recent years,
however, such a viewpoint has been changed with
the emergence of a number of methods/tools
which dedicate to schematically depicting chemical
and biological information on two-dimensional (2D)
diagrams. These simplified diagrams are aesthetic and
straightforward for the exhibition of molecular
configurations, profiles, behaviors and interactions.
Therefore, this area is considered as a branch of
molecular graphics, and here we call it 2D molecular
graphics. In fact, 2D and 3D visualization methods
are complementary. They are used for different
purposes and/or different occasions. For example,
a rotatable molecular stereograph provides a faithful
representation of available data but requires considerable operator time and practice to thoroughly
perceive and is often an inconvenient way to
communicate information among peers. A flattened
Corresponding author. Zhicai Shang, Department of Chemistry, Zhejiang University, Hangzhou 310027, China. Tel: þ86 571
87952379; Fax: þ86 571 87951895; E-mail: [email protected]
Peng Zhou is a PhD student at the Institute of Molecular Design & Molecular Thermodynamics, Department of Chemistry, Zhejiang
University, China. His research focus is in graphic design and analysis of chemical and biological systems.
Zhicai Shang is Professor of Department of Chemistry at the Zhejiang University, China. His research interests include the computeraided molecule design, bioinformatics and quantitative structure–activity relationship.
ß The Author 2009. Published by Oxford University Press. For Permissions, please email: [email protected]
248
Zhou and Shang
image can be designed to be very easily understood
even at a glance and can be readily shared as a printed
hardcopy.
OVERVIEW OF 2D MOLECULAR
GRAPHICS
Computer is a powerful tool to implement graphical
operations and its development has had a dramatic
impact upon 2D molecular graphics. It should always
be remembered, however, that there is much more
to 2D molecular graphics than the naissance of
computer. The history of 2D molecular graphics
could be traced back to the proposal of benzene’s
structure by Kekule [4]. After that, the chemical
structure became the most frequently used concept
in chemistry and the related fields. In recent decades,
2D molecular graphics was further developed and
applied in molecular biology. For example, protein
topology diagram and RNA secondary structure
diagram give a clear expression of the arrangement
of secondary structure elements and functionally
structural domains; the macromolecular surface
map can intuitively reflect the distribution situation
of residues on a protein surface. Nowadays, 2D
molecular graphics is also a useful tool in computerassisted drug design; pharmacologists can easily
identity and conveniently analyze pharmacophore
and nonbinding interaction of receptor–ligand complexes by using diagrammatic analysis applications.
These works suggest the 2D molecular graphics is
a promising branch of molecular graphics and will
be further developed rapidly in future. Here we
categorize 2D molecular graphics into three aspects
in terms of the research object and purpose:
(i) Generation and drawing of chemical structures.
(ii) Representation and annotation of biomolecules
(proteins and nucleic acids). (iii) Schematic layout
of molecular interactions.
CLASSIFIED DISCUSSION OF 2D
MOLECULAR GRAPHICS
Generation and drawing of chemical
structures
Generating 2D chemical structures from molecular
formulas, connection tables or spatial coordinates
belongs to the earliest application of computer in
chemistry. Nowadays, graphical drawing of chemical
structures is a relatively mature technique, readily
available in many commercial and academic software
such as ChemDraw [5], Symyx/Draw (formerly
named ISIS/Draw) [6], OGHAM [7], CACTVS [8],
etc. The demand for depiction of molecular structures is great, because the stylistic conventions
of such pictograms constitute a ‘natural language’
for chemists which can be rapidly comprehended,
and is thus highly suitable for communication.
Fortunately, chemical structures constitute a limited
subset of all possible graphs. If one considers only
organic chemistry, there can be formulated a
relatively small set of rules which cover most of
the common environments in which an atom type
might occur. Once these rules are stated, the
structure can be pieced together and an overall
depiction obtained [9].
Correct, aesthetic and understandable are the
basic requirement for generating chemical structures.
In the early stage, researchers mainly adopted some
simple constraints to generate chemical structures.
First attempt by Zimmerman and Feldman [10, 11]
used a library of basic substructures and combined
them according to a set of rules. Although this
method is particularly effective for drawing edgefused polycyclic ring systems such as steroids, it
is limited in its generality by the size and nature
of both the template set and the rule set, with
the most difficult cases being bridged polycyclic
systems. In the same time or later, several methods
were proposed for the similar purpose. However
these methods are all incapable of handling complex
molecules and thus do not to be applied widely
[12–14]. After 1980, intelligent algorithms were
introduced into this field, obtaining a good effect
on both computational efficiency and scale. Shelley
[15] presented a heuristic approach to avoid overlapping atoms and bonds and to facilitate similar
representations of the structural diagram for identical
or similar structures. Hibbert [16] employed genetic
algorithm to generate and display chemical structures, in which the atomic coordinates or bond
angles are encoded into chromosome and the
distances between all pairs of bonded and nonbonded atoms are served as the fitness function. The
DEPICT program written by Weininger [17] was
designed to convert SMILES format, the linear
notation of molecular graph, into a depiction of
molecular structure. The output display allows all
aspects of SMILES representation of structure to be
verified easily, including aromaticity, formal charge,
bond order assignment and hydrogen attachment.
A comprehensive review on the research status
2D molecular graphics
249
Figure 1: Some examples of 2D chemical structures generated by different algorithms. (a) Organometallic complex
(ChemDraw [5]). (b) Boride (CACTVS [8]). (c) Carbon nano-tube (Clark’s method [9]). (d) Cage-shaped carbon
molecule (OGHAM [7]).
of 1980s and 1990s for structure diagram generation
can be found in Ref. [18].
After 2000, a number of methods and programs
were proposed in elaborate consideration to generate, depict and edit 2D molecular structures. Their
application scope covers not only the common
organic compounds but also some special cases such
as organometallic complexes, borides, carbon nanotubes, cage-shaped carbon molecules, etc. (Figure 1).
Boissonnat et al. [19] firstly defined the concept
of molecular family that was a set of molecules
sharing a connected supertree. Using the correspondences between atoms provided by the supertree,
the drawing performing of molecular graphs can be
accurately coordinated by a breadth-first traversal
approach. Fricker et al. [20] further proposed an
algorithm that was based on the classical scheme of
a drawing queue placing the molecular fragments
in a sequential way. This method extended the
traditional prefabricated units developed for complex
ring systems to automatically created drawing units
for chains and rings which will then be assembled
in a sequential manner. Recently, Clark and
co-workers [9] have stated the two goals of a
successful structure depiction method: (i) to maximize the fraction of molecular connection graphs
that are readily discernible to the observer and (ii) to
maximize the fraction of graphs which are aesthetically ideal and functionally equivalent to the efforts
of a skilled artist. In addition, two works are deserved
to be mentioned: (i) one is the chiral depiction
algorithm by Maehr [21] who used the targeted
redeployment of established stereobonds to serve
as the stereodescriptors that can differentiate between
enantiopure and racemic compounds and those
that are enantiopure but whose chirality sense is
unknown; (ii) another is the small-molecule topology generator PRODRG [22] which takes input
from given coordinates or various 2D formats and
automatically generates molecular topologies suitable
for X-ray refinement of protein–ligand complexes.
On the other hand, two java-based programs,
JChemPaint [23] and MCDL editor [24], were
published for generating and editing chemical
structures. They are different from previous structure-generating programs in that their source codes
are shared and collaboratively developed through
the Internet, and these characteristics lead to that
the open source programs can be rapidly spread
and widely applied in the science community.
Representation and annotation
of proteins
Protein topology representation
Protein topology was originally investigated systemically by Crippen [25, 26] and gradually becoming
a common concept of protein structures. 2D
representation of protein topology structures is a
useful tool in protein searching [27] and fold
comparison [28, 29]. However, the main application
of protein topology representation is, of course, to
straightforwardly show the structural pattern and
spatial arrangement of protein architectures. Protein
topology representations can be classified into two
styles: cartoon diagram and topology diagram.
Cartoon diagram utilizes various symbols such
as circles, squares, triangles and lines to represent
protein’s substructures (Figure 2a), which was first
adopted by Schulz et al. [30] and Rossmann et al.
[31]. Levitt and Chothia [32] used a similar way to
illustrate the arrangement of -helices and -sheets
in 31 globular proteins with known conformations.
Based upon the representation results, they arranged
protein structures into four classes that are well
known today, i.e. all- proteins, all- proteins, þ proteins and / proteins. Cartoon diagrams were
later drawn manually in some publications to
elucidate the arrangement fashion of protein
250
Zhou and Shang
Figure 2: Stereoview of the Bacillus subtilis DppA monomer (center) (PDB entry: 1hi9). This figure was produced
using PyMOL [50]. (a) Cartoon diagram produced using TOPS [38]. (b) Topology diagram obtained from PDBsum
database [51].
structural elements [33, 34]. Flores et al. [35] have
described a general algorithm for automatically
generating protein cartoon diagrams, which was
further improved by Westhead et al. [36, 37] and
integrated into the TOPS database [38]. Recently,
May et al. [39] proposed a modified version of
protein cartoon diagram called PTGL in which
the secondary structure units were notated as small
circles and quadrates and the spatial neighbourhoods
were represented as arcs between these units.
Topology diagram consists of parallel/anti-parallel
arrows (for strands) and cylinders (for helices)
(Figure 2b). There were considerable publications
that adopted topology diagrams to illustrate protein
structural configurations [40–42], and several software tools were also available which take atomic
coordinate data input and produce protein topology
diagrams in different styles [43, 44]. The key
notation describing topology diagrams were first
introduced by Richardson [45] for -sheet topologies and further developed by Koch, Flower and
Grigoriev et al. [46–48]. The topology diagrams
are not only competent for clearly expressing
simple protein motifs, such as the Greek-key and the
Jelly-roll barrel motifs [49], but also performing well
for showing complex protein global structures.
Protein structure annotation
The aim of structure annotations is to present the
elusive structure-related information of proteins in a
straightforward diagram, in qualitative or quantitative
manner. Despite numerous protein diagrams were
drawn in publications by manual approaches or using
computer graphic software, the methods and programs used to automatically generate annotating
diagrams for proteins were also in lack, relative to its
extensive applications.
A normal style for protein structure annotations
is the linear representation which unfolds the
structured proteins into a linear state, and then uses
different patterns and colors to annotate structural
implications. One of the mostly used aspects of
linear representation is to show the arrangement of
secondary structures in protein configuration. This
is frequently adopted by database pages and publications [52]. Another important aspect of linear
representation is to display structural features from
sequence alignment [53, 54]. In addition, linear
representation has many other applications in
information presentation of protein structures at
sequence level, such as protein hydrophilicity plot
[55], flexibility plot [56] and structure validation
plot [57] (GCG [58] and PSAweb server [59] are
two useful tools to draw these plots based on protein
sequences and structures). Here we present three
interesting applications of linear representation:
(i) DOMPLOT [60] is a standalone program used
to generate line–box diagrams of protein chains in
terms of their structural domains. This program is
written in C and thus can be implemented
efficiently. (ii) Sipos et al. [61] developed a rapid
visual approach, pairwise repeat homology diagram,
to visualize the protein sequence repeats detected
by a profile HMM in a linear pair of sequences
and to highlight their homology relations inferred
by a phylogenetic tree. A Perl-script t2prhd was
also provided by them for drawing the diagrams.
(iii) Kolodny and Honig [62] proposed a method
called VISTAL that mapped the C-spatial distances
between two superposed proteins into a 2D
sequence alignment which described the superposed
structures as a series of matched secondary structure
elements, colored according to the distance of their
C atoms.
2D molecular graphics
Besides the linear representation, protein structure
annotation can also be performed using various
fashions for different purposes. Wako and Scheraga
[63] applied a distance constraint approach to 2D
modeling of proteins in order to visualize the nature
of protein folding and to examine the relative roles
of different ranges of interaction. Campagne et al.
[64–66] developed a web-based server RbDe that
greatly simplifies the construction of schematic
diagrams of proteins; the yielded residue-based
diagrams displayed the sequence of a given protein
in the context of its secondary and tertiary structure
and highlighted the amino acid residues in different
colors according to the physicochemical properties.
Kannan et al. [67] proposed a logo-based strategy
to express the information about neighbor composition densities for residues with different secondary
structures, and this method was used to ascertain
the reason why a particular sequence with a given
amino-acid composition should be fold into a
specific structural class. Furthermore, the solvent
accessibility of amino acids in structured proteins
is a critical property of proteins [68] and can be
calculated by many approaches such as MSMS [69],
NACCESS [70] and SAVOL3 [71]. However, there
was only one tool called ASAView [72, 73] that
can provide the schematic representation of accessible surface area for the independent amino acid
residues in buried state, by using spiral plot or
histogram.
Another important application of protein structure annotation is the sequence logo which was
originally described by Schneider and Stephens [74].
This method graphically represents the information
present in a multiple sequence alignment. Each
position in the logo is represented by a stack
consisting of letters occurring at the corresponding position in the alignment. The height of each
character is proportional to its frequency, where
the most common character is placed at the top
of the stack. The height of the entire stack indicates
the information content available at that position.
Currently, several tools for creating protein sequence
logos are available via the web interfaces [75–77].
Sequence logo can also be used for DNA and RNA.
Intra-protein nonbonding/bonding plot
The global structure of proteins is stabilized by
hydrophobic packing occurring in the cores, whereas
the conformational specificity is accurately adjusted
only by the forming or breaking of the bonding
251
(e.g. disulphide bridges) and nonbinding (e.g.
hydrogen bonds) interactions between intra-protein
elements. It is conceivable that the complicated
interaction network of protein structures is difficult
to be understood in the stereoviews. Therefore,
several graph theory-based methods were proposed
to visualize the disulphide bridges and hydrogen
bond networks of proteins in the 2D pages.
2D visualization of hydrogen bond networks.
HERA [78] is the first program for automatically
plotting hydrogen bonding diagrams of protein
structures. It can also be used to calculate the
connectivity of -strands and to extract simple
structural motifs such as hairpins or Greek keys.
Another similar tool is HBNG [79] which gives
an enumeration of favorable topologies of hydrogen
bond networks in protein structures and determines
the effect of cooperativity and anticooperativity on
protein stability and folding.
2D visualization of disulphide bridges. Protein
disulphide bridge topology has been investigated
intensively [80–83] and is usually schematically
represented by topology pattern plot. The topological stereoisomerism of multiple-disulfide polypeptide chain in stable folded proteins was first
characterized from graph theoretic analysis of its
covalent structure by Mao [84]. Based on that,
Benham and Jafri [85] made a systematical study
on disulfide bonding patterns and defined a simple
linear graph to illustrate the linking patterns of
disulfide bridges in protein sequences. The similar
way was also used to analyze and visualize the
topological stereochemistry of protein hydrogen
bonds [86].
Protein surface map
The original work of projecting biological surface
onto a 2D plane was made by Rossmann and
Palmenberg [87] who used the ‘roadmaps’ to depict
the residue distribution on the capsids of human
rhinovirus and mengovirus. Chapman [88] extended
the application scope of ‘roadmaps’ to the external
surfaces of proteins and gave a clear definition of the
boundaries between labeled residues in the maps.
Recently, researcher’s interest has been transferred
from protein global surfaces to some concerned local
regions. Stahl et al. [89] proposed a computer-based
strategy for automatically mapping protein surface
cavities. Using this approach they successfully
examined 176 metalloproteinases containing zinc
cations in the protein active sites. Lee and Tzou [90]
252
Zhou and Shang
described an earth-map representation of the directionality of protein surfaces in the DNA-binding
regions. Gabdoulline et al. [91] developed a tool
MolSurfer for analyzing, characterizing and visualizing the physicochemical properties of protein–
protein-binding interfaces in 2D pages.
Transmembrane protein display
Transmembrane proteins play many important and
diverse functional roles in cells and biological
processes [92]. A number of methods and tools
were reported for producing 2D transmembrane
diagrams from protein primary sequences or
advanced structures, such as the above mentioned
RbDe [93] which can also be used to generate
residue-based schematic diagram of transmembrane
proteins. Moreover, the G protein-coupled receptor
database (GPCRDB) [94] provides a 2D visualization tool Viseur [95] to display snake like diagram for
the queried GPCRs. VHMPT [96] is a transmembrane topology plotting program that allows users
to modify the layout of the generated topology, to
label specific amino acid or amino-acid groups,
and to annotate with arrows and texts. TMRPres2D
[97] automates the generation of uniform, high
analysis 2D graphical images of -helical or -barrel
transmembrane proteins. TOPO2 [98] is a simple
graphics program using a sequence along with
input provided by the user to produce an image of
a transmembrane protein. TEXtopo [99] creates the
transmembrane topologies from PHD predictions
[100] or SwissProt database files [101], and the
output diagrams use shades and rich decorations to
emphasize conserved residues or functional properties of the residue side-chains.
Representation and annotation
of nucleic acids
2D representation of DNA
The mostly used 2D representation of DNA is, just
like proteins, the linear representation of which the
main applications are concentrated on sequence
characteristics display and alignment plotting, these
functions are available in many software suites such as
BioViews [102], viewGene [103], SeqVISTA [104],
etc. DNA sequence logo is also a popular fashion
for annotating nucleotide’s conservation or other
properties [105]. For example, the Sequencewalker
[106] is a special tool adopting logo style to display
how binding proteins and other macromolecules
interact with individual bases of a given nucleotide
sequences. Another important aspect of DNA 2D
representation is the random walk plot in which the
four nucleotides A/T and C/G are served as movers
indicating respectively left/right and up/down
moves, a DNA sequence can thus ‘walk’ on a 2D
plane, from 50 to 30 , to form a polyline plot [107–109].
However, the main uses of DNA walk plot are in
sequence characterization [110] and analysis [111],
rather than the 2D visualization. Moreover, there has
been a simple but interesting instance of visualizing
complete genomes, called DNA rainbow [112]. This
method is a beautiful mix of science and art, which
assigns a color to every of the four bases in a chromosome and renders it as a multicoloured picture.
In a previous review, Roy et al. [113] have
given a detail description of the DNA graphical
representation.
2D representation of RNA
The focus of RNA 2D representation is how to
generate non-overlapping arrangement and layout
of RNA secondary structure elements under an
aesthetically pleasing consideration. In a mathematical sense, the secondary structure of RNA is a
topological structure rather than a geometric structure. It remains unchanged even when under
distortion, as long as the connectivity relation
between nucleotides is not changed. Han et al.
[114] have classified the RNA 2D representation
methods into conventional and unconventional
representations. The published works on 2D representation of RNA are listed in Table S1 in
Supplementary Data.
Conventional representation of RNA. The conventional representation illustrates a polygonal display
of RNA global structure in which loops are drawn as
regular polygons, thereby generating clear and compact representation, especially for the long RNAs.
In the early stage, Lapalme et al. [115] presented a
simple algorithm for short RNA segments. Shapiro
et al. [116] later developed a method to circumvent
the problem of overlapping portions of the RNA.
Bruccoleri and Heinrich [117] described an improved
algorithm that is particularly suitable for drawing large
RNA molecules with very limited overlap of strands.
Gautheret et al. [118] gave a brief review on the
early studies of RNA display. In the 1990s, RNA 2D
representation obtained a rapid development in
both methodology and software. These works laid
emphasis on different aspects ranging from the
generation of overlap-free polygonal displays [119]
2D molecular graphics
253
and richly annotated diagrams [120] to development
of general graphic libraries [121] and efficient
computer algorithms [122]. After 2000, RNA
conventional representation has gradually grown to
be mature. Recently, it trends to developing combinatorial application suites that assemble a number of
modules with diverse functions to automatically
implement the complete procedure of RNA representation, including sequence alignment, energy
optimization, secondary prediction, and so on.
Unconventional representation of RNA. The unconventional representation pays attention to local characteristics of RNA structures (e.g. loops, domes,
pseudoknots, base pairs, etc.) or uses special fashions
to define RNA global structures. Compared to
conventional representation, methods and tools for
unconventional representation are relatively few but
developing rapidly in recent years. Martinez [123]
early proposed a multiple approach for identifying
and depicting local structures in the form of RNA
hairpins. After that, Han et al. [124] presented an
algorithm drawing H-type pseudoknots based on
RNA graph theory, and this method was further
improved in several aspects including pseudoknot
types, display quality and computational efficiency
[125]. Another important tool in the RNA unconventional representation area is the RNAView [126],
which creates a schematically diagram for all base
pairs in a RNA molecule using previously developed
theory and symbols by the same group [127, 128].
RNAView was further modified to Rna2DViewer
and integrated into the S2S package [129]. Like the
RNAView, MC-Annotate [130] can also be used
to annotate information about RNA local structures.
The output of this program is composed of a 2D
structural graph that contains the annotations, and
a series of HTML documents, one for each nucleotide conformation and base–base interaction present
in the input 3D structure. Moreover, the tool of
jViz.Rna developed by Wiese et al. [131] is a Java
program that represents RNA global structures using
various strange patterns, such as linked graph, circle
graph, dot plot, etc. In this literature they also gave a
brief description of some useful RNA visualization
tools.
enzyme–substrate, receptor–ligand, target–drug, etc.
Knowledge of these interactions is important to
understanding how proteins function in biological
systems and to developing drugs with specific
selectivity. 2D representation of protein–small molecule interactions dedicates to straightforwardly
demonstrating the hidden active sites and the
complicated nonbonding networks across the binding interfaces. However, only three tools are
available that can be used to automatically generate
schematic 2D diagrams of protein–small molecule
interactions.
LIGPLOT is a well-known program written by
Wallace et al. [132]. It employs HBPLUS [133] and
NACCESS [70] to identify and compute hydrogen
bonds and solvent accessible areas around the ligand
molecule, and subsequently arranging these information into a neat page containing a complete
description of the nonbinding effects on the given
complex’s interface (Figure 3a). LIGPLOT adopts
a series of algorithms to avoid overlapping and
to flatten ring groups, and provides diverse output
styles that can be easily changed by users.
In the MOE platform, Clark and Labute [134]
employed scientific vector language (SVL) to fulfill
the procedure of automatic depiction of protein–
ligand complexes. The output diagrams are annotated with a substitution contour, solvent exposure,
cofacters, covalently bound linkages and a series of
identified nonbonding interactions and conserved
residues (Figure 3b). This method can also be applied
to aligned sets which contain multiple ligands, or
multiple members of a protein family, in which case
the ligand orientations and protein residue placement
will show consistent trends throughout the series.
Stierand et al. [135, 136] described an algorithm
that is based on a combinatorial optimization strategy
which can solve parts of the layout overlapping
problem nonheuristically. The depicted molecules
are represented as structure diagrams according
to their chemical nomenclatures and arranged to
maximize aesthetic ideals (Figure 3c). This method is
called PoseView and available as a web server by
which the complete protocol of layout procedure
can be implemented efficiently and conveniently.
Schematic layout of molecular
interactions
Protein^nucleic acid interaction
Today, only a single software tool for the automatic
creation of protein–nucleic acid complex diagrams
called NUCPLOT [138] is available. It is a
standalone program that can automatically identify
Protein^small molecule interaction
Protein–small molecule interactions are accompanied
with the binding and association processes of
254
Zhou and Shang
Figure 3: 2D representation of the cyclin-dependent kinase 2 (CDK2)-oxindole inhibitor complex (PDB entry: 1fvt)
(below).These diagrams were generated using (a) LIGPLOT [132], (b) MOE [134] and (c) PoseView [135].The stereoview
of the complex (above) was produced using CHIMERA [137].
interfacial residues and interactions between them
from the input PDB coordinates of the complex, and
generates a plot in which all interaction information
are intuitively shown in a schematic manner. The
program creates the output files consisting of nucleic
acid chains, interfacial elements, hydrogen bonds,
van der Waal contacts and covalent bonds between
the protein and the nucleic acid. The resulting diagram is both clear and simple and allows immediate
identification of important interactions within the
structure. The program works well for both doublestranded and single-stranded nucleic acids binding
with proteins, so it can be used for both cases of
protein–DNA complexes and partial protein–RNA
complexes.
Protein^protein interaction
The ready availability of structural data of protein
complexes, both from experimental determination,
such as by X-ray crystallography, and by theoretical
modeling, such as protein docking, has made it
necessary to find ways to easily interpret the results
at 2D level. The simplest way used for 2D representation of protein–protein interactions is the
linking plot in which the protein sequences are
represented as two lines, each from one protein
subunit of the complex, and the residue-pairs in
interactions are linked linearly. Several methods
like MONSTER [139] and iPfam [140] can be
used to automatically generate linking plots from
protein complex structures. The main disadvantage
of linking plot is that the presenting information
is too cursory to give a comprehensive insight into
the protein-binding interfaces. For that, QContacts
[141] adopts a different strategy called contacting
map to exhibit nonbonding interactions in the
binding interfaces. This map can be viewed as a
matrix; residues in turn from one subunit sequence
define the columns, and those from another define
the rows, the matrix elements thus represent the
corresponding residue-pairs and, if in interaction,
labeled by different symbols indicating different
interaction types. A more aesthetically pleasing,
informative diagram showing protein–protein interactions can be generated by using DIMPLOT [132],
which is a sister program of above-mentioned
LIGPLOT and thus its function and output style
are same to that of LIGPLOT.
Recently, our group developed a graphical user
interface (GUI)-based package 2D-GraLab [142] for
2D molecular graphics
automatically generating comprehensive representation of diverse nonbonding (and covalent) interactions across the protein-binding interfaces, including
hydrogen bonds, salt bridges, van der Waals interactions, hydrophobic contacts, – stackings,
disulfide bonds, desolvation effect and loss of conformational entropy. This program takes standard
PDB format as the input file and gives two types
of PostScript outputs, referring to individual schematic diagram (for each nonbinding type) and
summarized schematic diagram (for all nonbinding
types). In addition, 2D-GraLab provides a complete
protocol to define and calculate geometric and energetics parameters for different nonbonding types,
such as length, angle, free energy, stability, etc., and
subsequently presents these parameters with interfacial structure elements in an elaborately designed
page. To ensure these interactions are determined
accurately and reliably, methods and standalone
programs employed in 2D-GraLab are all widely
used in the chemistry and biology communities.
CONCLUSIONS
2D molecular graphics is the subject that uses an
intuitive, understandable and accurate manner to
present configuration and information of chemical
and biological systems in planar pages. It should be
always noted that the two points: (i) The data and
parameters used by 2D molecular graphics are all
taken from 3D structures and models, and further
mapped into 2D representations by a consensus
algorithm. Therefore, the diagrams of 2D molecular
graphics are different from those of inexact drafted
pictures that are used just for schematic demonstration. (ii) 2D molecular graphics lays emphasis on
graphic representation, rather than being severed
as calculation tool. Therefore, those studies using
2D graph and geometry just for the purposes of
mathematical tool or modeling approach, such as
chemical graph theory and DNA topological theory,
do not belong to the spectrum of 2D molecular
graphics.
Key Points
Propose the term of 2D molecular graphics.
Review methods, tools and applications of 2D molecular
graphics.
Define the difference between the 2D molecular graphics and
other similar subjects.
255
Acknowledgements
The authors would like to express their gratitude to the editors
and the anonymous reviewers for their professional, intensive
and useful comments. They also thank the people for providing
access of pictures, programs, servers and databases used in this
article.
SUPPLEMENTARY DATA
Supplementary data are available online at http://
bib.oxfordjournals.org/.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Ernesto Z. TheTheory of Graphs in Linguistics. The Hague:
Mouton Press, 1970.
Goodsell DS. Visual methods from atoms to cells. Structure
2005;13:347–54.
Leach AR. Molecular Modelling: Principles and Applications
(2nd Ed). UK: Pearson Education EMA, 2001.
Bensaude-Vincent B, Stengers I. A History of Chemistry.
Cambridge: Harvard University Press, 1997.
ChemDraw. CambridgeSoft Corp., MA, USA.
Symyx/Draw. Symyx Technologies, Inc., CA, USA.
Ogham. OpenEye Scientific Software, Inc., NM, USA.
Ihlenfeldt WD, Takahashi Y, Abe H, et al. Computation
and management of chemical-properties in CACTVS:
an extensible networked approach toward modularity and
compatibility. J Chem Inf Comput Sci 1994;34:109–16.
Clark AM, Labute P, Santavy M. 2D structure depiction.
J Chem Inf Model 2006;46:1107–23.
Zimmerman BL. Computer-Generated Chemical Structural
Formulas with Standard Ring Orientations, PhD Dissertation,
University of Pennsylvania, 1971.
Wipke WT, Heller S, Feldmann R, et al. Computer
Representation and Manipulation of Chemical Information.
New York: Wiley, 1974.
Carhart RE. A model-based approach to the teletype
printing of chemical structures.. J Chem Inf Comput Sci 1976;
16:82–8.
Dittmar PG, Mockus J, Couvreur KM. An algorithmic
computer graphics program for generating chemical structure diagrams. J Chem Inf Comput Sci 1977;17:186–92.
Blake JE, Farmer NA, Haines RC. An interactive computer
graphics system for processing chemical structure diagrams.
J Chem Inf Comput Sci 1977;17:223–8.
Shelley CA. Heuristic approach for displaying chemical
structures. J Chem Inf Comput Sci 1983;23:61–5.
Hibbert DB. Generation and display of chemical structures
by genetic algorithms. Chemometr Intel Lab Syst 1993;20:
35–43.
Weininger D. SMILES. 3. DEPICT. Graphical depiction of
chemical structures. J Chem Inf Comput Sci 1990;30:237–43.
Helson HE. Structure diagram generation. Rev Comput
Chem 1999;13:313–98.
Boissonnat JD, Cazals F, Flötotto J. 2D-structure drawings
of similar molecules. Proc 8th Int Symp Graph Draw 2000;
115–126.
256
Zhou and Shang
20. Fricker PC, Gastreich M, Rarey M. Automated drawing
of structural molecular formulas under constraints. J Chem
Inf Comput Sci 2004;44:1065–78.
21. Maehr H. Graphic representation of configuration in
two-dimensional space. current conventions, clarifications,
and proposed extensions. J Chem Inf Comput Sci 2002;42:
894–902.
22. Schüttelkopf AW, van Aalten DMF. PRODRG: a tool for
high-throughput crystallography of protein-ligand complexes. Acta Cryst 2004;D60:1355–63.
23. Krause S, Willighagen E, Steinbeck C. JChemPaint, using
the collaborative forces of the internet to develop a free
editor for 2D chemical structures. Molecules 2000;5:93–8.
24. Trepalin SV, Yarkov AV, Pletnev IV, et al. A Java chemical
structure editor supporting the modular chemical descriptor
language (MCDL). Molecules 2006;11:219–231.
25. Crippen GM. Topology of globular proteins. J Theor Biol
1974;45:327–38.
26. Crippen GM. Topology of globular proteins. II. JTheor Biol
1975;51:495–500.
27. Gilbert D, Westhead D, Nagano N, et al. Motif-based
searching in TOPS protein topology databases. Bioinformatics
1999;15:317–26.
28. Martin AC. The ups and downs of protein topology; rapid
comparison of protein structure. Protein Eng 2000;13:
829–37.
29. Johannissen LO, Taylor WR. Protein fold comparison by
the alignment of topological strings. Protein Eng 2003;12:
949–55.
30. Schulz GE, Schirmer RH. Topological comparison
of adenyl kinase with other proteins. Nature 1974;250:
142–4.
31. Rossmann MG, Moras D, Olsen KW. Chemical and
biological evolution of a nucleotide-binding protein. Nature
1974;250:194–9.
32. Levitt M, Chothia C. Structural patterns in globular
proteins. Nature 1976;261:552–7.
33. Naganoa K. Logical analysis of the mechanism of protein
folding. IV. Super-secondary structures. J Mol Biol 1977;109:
235–50.
34. Sternberga MJE, Thorntona JM. On the conformation of
proteins: the handedness of the connection between parallel
-strands. J Mol Biol 1977;110:269–83.
35. Flores TP, Moss DS, Thornton JM. An algorithm for
automatically generating protein topology cartoons. Protein
Eng 1994;7:31–7.
36. Westhead DR, Hatton DC, Thornton JM. An atlas of
protein topology cartoons available over the world wide
web. Trends Biochem Sci 1998;23:35–6.
37. Westhead DR, Slidel TW, Flores TP, et al. Protein
structural topology: automated analysis and diagrammatic
representation. Protein Sci 1999;8:897–904.
38. Michalopoulos I, Torrance GM, Gilbert DR, et al. TOPS:
an enhanced database of protein structural topology. Nucleic
Acids Res 2004;32:D251–4.
39. May P, Barthel S, Koch I. PTGL—a web-based database
application for protein topologies. Bioinformatics 2004;20:
3277–9.
40. Jayaram H, Taraporewala Z, Patton JT, et al. Rotavirus
protein involved in genome replication and packaging
exhibits a HIT-like fold. Nature 2002;417:311–5.
41. Bewley MC, Graziano V, Jiang J, et al. Structures of
wild-type and mutant human spermidine/spermine
N1-acetyltransferase, a potential therapeutic drug target.
Proc Natl Acad Sci USA 2006;103:2063–8.
42. He X, Zhou J, Bartlam M, et al. Crystal structure of the
polymerase PAC–PB1N complex from an avian influenza
H5N1 virus. Nature 2008;454:1123–6.
43. Collaborative Computational Project. The CCP4 suite:
programs for protein crystallography. Acta Cryst 1994;D50:
760–3.
44. Bond CS. TopDraw: a sketchpad for protein structure
topology cartoons. Bioinformatics 2003;19:311–2.
45. Richardson JS. -sheet topology and the relatedness of
proteins. Nature 1977;268:495–500.
46. Koch I, Kaden F, Selbig J. Analysis of protein sheet
topologies by graph-theoretical methods. Proteins 1992;12:
314–23.
47. Flower DR. -sheet topology. A new system of nomenclature. FEBS Lett 1994;344:247–50.
48. Grigoriev I, Mironov A, Rakhmaninova A. Inter-helical
contacts determining the architecture of -helical globular
proteins. J Biomol Struct Dyn 1994;12:559–72.
49. Zhang C, Kim S-H. A comprehensive analysis of the
Greek key motifs in protein -barrels and -sandwiches 2000;40:
409–19.
50. DeLano WL (2002), The PyMOL Molecular Graphics
System. DeLano Scientific, San Carlos, CA, USA.
51. Laskowski RA, Hutchinson EG, Michie AD, et al.
PDBsum: a web-based database of summaries and
analyses of all PDB structures. Trends Biochem Sci 1997;22:
488–90.
52. Luscombe NM, Laskowski RA, Westhead DR, et al. New
tools and resources for analysing protein structures and their
interactions. Acta Cryst 1998;D54:1132–8.
53. Delamarche C. Color and graphic display (CGD): Programs
for multiple Sequence alignment analysis in spreadsheet
software. BioTechniques 2000;29:100–7.
54. Li W, Godzik A. VISSA: A program to visualize structural
features from structure sequence alignment. Bioinformatics
2006;22:887–8.
55. Kyte J, Doolittle RF. A simple method for displaying
the hydrophathic character of a protein. J Mol Biol 1982;157:
105–32.
56. Ragone R, Facchiano F, Facchiano A, et al. Flexibility plot
of proteins. Protein Eng 1989;2:497–504.
57. Laskowski RA, MacArthur MW, Moss DS, et al.
PROCHECK—a program to check the stereochemical
quality of protein structures. J Appl Crystallogr 1993;26:
283–91.
58. Genetics Computer Group, Science Drive, Madison, WI,
USA.
59. Raghava GPS. A graphical web server for the analysis
of protein sequences and alignment. Biotech Software Internet
Rep 2001;2:255–8.
60. Todd AE, Orengo CA, Thornton JM. DOMPLOT: a
program to generate schematic diagrams of the structural
domain organization within proteins, annotated by ligand
contacts. Protein Eng 1999;12:375–9.
61. Sipos B, Somogyi K, Andó I, et al. t2prhd: a tool to
study the patterns of repeat evolution. BMC Bioinformatics
2008;9:27.
2D molecular graphics
62. Kolodny R, Honig B. VISTAL—a new 2D visualization
tool of protein 3D structural alignments. Bioinformatics 2006;
22:2166–7.
63. Wako H, Scheraga HA. Visualization of the nature of
protein folding by a study of a distance constraint approach
in two-dimensional models. Biopolymers 1982;21:611–32.
64. Konvicka K, Campagne F, Weinstein H. Interactive
construction of residue-based diagrams of proteins: the
RbDe web service. Protein Eng 2000;13:395–6.
65. Campagne F, Bettler E, Vriend G, et al. Batch mode
generation of residue-based diagrams of proteins.
Bioinformatics 2003;19:1854–5.
66. Skrabanek L, Campagne F, Weinstein H. Building protein
diagrams on the web with the residue-based diagram editor
RbDe. Nucleic Acids Res 2003;31:3856–8.
67. Kannan N, Schneider TD, Vishveshwara S. Logos
for amino-acid preferences in different backbone packing
density regions of protein structural classes. Acta Cryst 2000;
D56:1156–65.
68. Connolly ML. Solvent-accessible surfaces of proteins and
nucleic-acids. Science 1983;221:709–13.
69. Sanner MF, Olson AJ, Spehner JC. Reduced surface: an
efficient way to compute molecular surfaces. Biopolymers
1996;38:305–20.
70. Hubbard SJ, Thornton, JM. NACCESS. Department of
Biochemistry and Molecular Biology, University College
London, UK.
71. Pearlman RS, Skell JM. SAVOL3. College of Pharmacy,
University of Texas, Austin, TX, USA.
72. Ahmad S, Wang J-Y, Gromiha MM, et al. Development
of residue look-up tables and graphical representation of
solvent accessibility in proteins. Genome Informatics 2003;14:
482–3.
73. Ahmad S, Gromiha M, Fawareh H, et al. ASAView:
database and tool for solvent accessibility representation in
proteins. BMC Bioinformatics 2004;5:51.
74. Schneider TD, Stephens MR. Sequence logos: a new way
to display consensus sequences. Nucleic Acids Res 1990;18:
6097–100.
75. Henikoff S, Henikoff JG, Alford WJ, et al. Automated
construction and graphical presentation of protein blocks
from unaligned sequences. Gene 1995;163:GC17–26.
76. Crooks GE, Hon G, Chandonia J-M, et al. WebLogo: a
sequence logo generator. Genome Res 2004;14:1188–90.
77. Pérez-Bercoff Å, Koch J, Bürglin TR. LogoBar: bar graph
visualization of protein logos with gaps. Bioinformatics 2006;
22:112–4.
78. Hutchinson EG, Thornton JM. HERA—a program to
draw schematic diagrams of protein secondary structures.
Proteins 1990;8:203–12.
79. Tiwari A, Tiwari V. HBNG: graph theory based visualization of hydrogen bond networks in protein structures.
Bioinformation 2007;2:28–30.
80. Kikuchi T, Nemethy G, Scheraga H. Spatial geometric
arrangements of disulfide-crosslinked loops in proteins.
J Comput Chem 1986;7:61–88.
81. Kikuchi T, Nemethy G, Scheraga H. Spatial geometric
arrangements of disulfide-crosslinked loops in non-planar
proteins. J Comput Chem 1989;10:287–94.
82. Boisbouvier J, Blackledge M, Sollier A, et al. Simultaneous
determination of disulphide bridge topology and
257
three-dimensional structure using ambiguous intersulphur
distance restraints: Possibilities and limitations. J Biomol
NMR 2000;16:197–208.
83. Li X, Zou G, Yuan W, et al. Defining the native disulfide
topology in the somatomedin B domain of human
vitronectin. J Biol Chem 2007;282:5318–26.
84. Mao B. Molecular topology of multiple-disulfide polypeptide chains. J Am Chem Soc 1989;111:6132–6.
85. Benham CJ, Jafri MS. Disulfide bonding patterns and
protein topologies. Protein Sci 1993;2:41–54.
86. Mao B, Chou K, Maggiora G. Topological analysis of
hydrogen bonding in protein structure. EurJ Biochem 1990;
188:361–5.
87. Rossmann MG, Palmenberg AC. Conservation of the
putative receptor attachment site in picornaviruses. Virology
1988;164:373–82.
88. Chapman MS. Mapping the surface properties of macromolecules. Protein Sci 1993;2:459–69.
89. Stahl M, Taroni C, Schneider G. Mapping of protein
surface cavities and prediction of enzyme class by a selforganizing neural network. Protein Eng 2000;13:83–8.
90. Lee W-P, Tzou W-S. Molecular surface directionality
of the DNA-binding protein surface on the earth map.
Genet Mol Biol 2006;29:408–12.
91. Gabdoulline RR, Wade RC, Walther D. MolSurfer: a
macromolecular interface navigator. Nucleic Acids Res 2003;
31:3349–51.
92. Byrne B, Iwata S. Membrane protein complexes. Curr
Opin Struct Biol 2002;12:239–43.
93. Campagne F, Weinstein H. Schematic representation of
residue-based protein context-dependent data: an application to transmembrane proteins. J Mol Graph Model 1999;17:
207–13.
94. Horn F, Weare J, Beukers MW, et al. GPCRDB: an
information system for G protein-coupled receptors. Nucleic
Acids Res 1998;26:275–9.
95. Campagne F, Jestin R, Reversat JL, et al. Visualisation
and integration of G protein-coupled receptor related
information help the modelling: description and applications of the Viseur program. J Comput Aid Mol Des 1999;13:
625–43.
96. Lin W-J, Hwang M-J. VHMPT: a graphical viewer
and editor for helical membrane protein topologies.
Bioinformatics 1998;14:866–8.
97. Spyropoulos IC, Liakopoulos TD, Bagos PG, et al.
TMRPres2D: high quality visual representation of
transmembrane protein models. Bioinformatics 2004;20:
3258–60.
98. Johns SJ. TOPO2. University of California. San Francisco,
USA.
99. Beitz E. TEXtopo: shaded membrane protein topology
plots in LATEX2". Bioinformatics 2000;16:1050–1.
100. Rost B. PHD: prediction one-dimensional protein
structure by profile based neural networks. Meth Enzymol
1996;266:525–39.
101. Bairoch A, Apweiler R. The SWISS-PROT protein
sequence data bank and its supplement TREMBL. Nucleic
Acids Res 1997;25:31–6.
102. Helt GA, Lewis S, Loraine AE, et al. BioViews: Java-based
tools for genomic data visualization. Genome Res 1998;8:
291–305.
258
Zhou and Shang
103. Kashuk C, SenGupta S, Eichler E, et al. viewGene: a
graphical tool for polymorphism visualization and characterization. Genome Res 2002;12:333–8.
104. Hu Z, Frith M, Niu T, et al. SeqVISTA: a graphical tool for
sequence feature visualization and comparison. BMC
Bioinformatics 2003;4:1.
105. Workman CT, Yin Y, Corcoran DL, et al. enoLOGOS:
a versatile web tool for energy normalized sequence logos.
Nucleic Acids Res 2005;33:W389–92.
106. Schneider TD. Sequence walkers: a graphical method to
display how binding proteins interact with DNA or RNA
sequences. Nucleic Acids Res 1997;25:4408–15.
107. Leong PM, Morgenthaler S. Random walk and gap plots
of DNA sequences. Comput Appl Biosci 1995;11:503–7.
108. Berger JA, Mitra SK, Carli M, et al. Visualization and
analysis of DNA sequences using DNA walks. J Franklin
Institute 2004;341:37–53.
109. Kim MA, Lee EJ, Cho HG, et al. A visualization technique
for DNA walk plot using -convex hull. Proc Fifth Int Conf
Central Europe Comput GraphVisualization 1997;10–14.
110. Randić R, Vračko M, Nandy A, et al. On 3-D graphical
representation of DNA primary sequences and their
numerical characterization. J Chem Inf Comput Sci 2000;40:
1235–44.
111. Thomas JM, Horspool D, Brown G, et al. GraphDNA:
a Java program for graphical display of DNA composition
analyses. BMC Bioinformatics 2007;8:21.
112. Bierkandt K, Bierkandt J. DNA rainbow. [http://
www.dna-rainbow.org] (21 March 2008, date last accessed).
113. Roy A, Raychaudhury C, Nandy A. Novel techniques of
graphical representation and analysis of DNA sequences—
a review. J Biosci 1998;23:55–65.
114. Han K, Kim D, Kim H-J. A vector-based method for
drawing RNA secondary structure. Bioinformatics 1999;15:
286–97.
115. Lapalme G, Cedergren RJ, Sankoff D. An algorithm for
the display of nucleic acid secondary structure. Nucleic Acids
Res 1982;20:8351–6.
116. Shapiro BA, Maizel J, Lipkin LE, et al. Generating nonoverlapping displays of nucleic acid secondary structure.
Nucleic Acids Res 1984;12:75–88.
117. Bruccoleri RE, Heinrich G. An improved algorithm for
nucleic acid secondary structure display. Comput Appl Biosci
1988;4:167–73.
118. Gautheret D, Major F, Cedergren R. Computer modeling
and display of RNA secondary and tertiary structures.
Methods Enzymol 1990;183:318–30.
119. Perochon-Dorisse J, Chetouani F, Aurel A, et al. RNA_d2:
a computer program for editing and display of RNA
secondary structures. Comput Appl Biosci 1995;11:101–9.
120. Chetouani F, Monestié P, Thébault P, et al. ESSA: an
integrated and interactive computer tool for analysing RNA
secondary structure. Nucleic Acids Res 1997;25:3514–22.
121. Muller G, Gaspin C, Etienne A, et al. Automatic display
of RNA secondary structures. Comput Appl Biosci 1993;9:
551–61.
122. Auber D, Delest M, Domenger JP, et al. Efficient drawing
of RNA secondary structure. J Graph Algorithms Appl 2006;
10:329–51.
123. Martinez HM. An RNA secondary structure workbench.
Nucleic Acids Res 1988;16:1789–98.
124. Han K, Lee Y, Kim W. PseudoViewer: automatic
visualization of RNA pseudoknots. Bioinformatics 2002;18:
S321–8.
125. Han K, Byun Y. PseudoViewer2: visualization of
RNA pseudoknots of any type. Nucleic Acids Res 2003;31:
3432–40.
126. Yang H, Jossinet F, Leontis N, et al. Tools for the automatic
identification and classification of RNA base pairs. Nucleic
Acids Res 2003;31:3450–60.
127. Leontis N, Westhof E. Geometric nomenclature and
classification of RNA base pairs. RNA 2001;7:499–512.
128. Leontis N, Westhof E. The annotation of RNA motifs.
Comp Funct Genom 2002;3:518–24.
129. Jossinet F, Westhof E. Sequence to structure (S2S): display,
manipulate and interconnect RNA data from sequence to
structure. Bioinformatics 2005;21:3320–1.
130. Gendron P, Lemieux S, Major F. Quantitative analysis
of nucleic acid three-dimensional structures. J Mol Biol 2001;
308:919–36.
131. Wiese KC, Glen E, Vasudevan A. jViz.Rna—a Java tool
for RNA secondary structure visualization. IEEE Trans
Nanobiosci 2005;4:212–8.
132. Wallace AC, Laskowski RA, Thornton JM. LIGPLOT: a
program to generate schematic diagrams of protein-ligand
interactions. Protein Eng 1995;8:127–34.
133. McDonald IK, Thornton JM. Satisfying hydrogen bonding
potential in proteins. J Mol Biol 1994;238:777–93.
134. Clark AM, Labute P. 2D depiction of protein-ligand
complexes. J Chem Inf Model 2007;47:1933–44.
135. Stierand K, Maaß PC, Rarey M. Molecular complexes at
a glance: automated generation of two-dimensional complex diagrams. Bioinformatics 2006;22:1710–6.
136. Stierand K, Rarey M. From modeling to medicinal
chemistry: automatic generation of two-dimensional complex diagrams. ChemMedChem 2007;2:853–60.
137. Pettersen EF, Goddard TD, Huang CC, et al. UCSF
Chimera – A visualization system for exploratory research
and analysis. J Comput Chem 2004;25:1605–12.
138. Luscombe NM, Laskowski RA, Thornton JM.
NUCPLOT: a program to generate schematic diagrams of
protein–nucleic acid interactions. Nucleic Acids Res 1997;25:
4940–5.
139. Salerno WJ, Seaver SM, Armstrong BR, et al. MONSTER:
inferring non-covalent interactions in macromolecular
structures from atomic coordinate data. Nucleic Acids Res
2004;32:W566–8.
140. Finn RD, Marshall M, Bateman A. iPfam: visualization of
protein–protein interactions in PDB at domain and amino
acid resolutions. Bioinformatics 2005;21:410–2.
141. Fischer TB, Holmes JB, Miller IR, et al. Assessing methods
for identifying pair-wise atomic contacts across binding
interfaces. J Struct Biol 2006;153:103–12.
142. Zhou P, Tian F, Shang Z. 2D depiction of nonbonding
interactions for protein complexes. J Comput Chem,
2009;30:940–51.