Purpose:FirstGlance in Jmol is the easiest way to look at the 3D
structures of proteins, DNA, RNA, and their complexes. It works
within a web browser, and was designed to enable the readers of
scientific journals to see the main features of newly published
3D models in a few clicks, without installing anything, and in
all popular web browsers and computer platforms. It was adopted by
Nature (2007-2010),
and is currently (January 2013) in use at
Nature Structural and Molecular Biology
(see their 3D View links),
Proteopedia.Org,
the
Protein Data Bank,
the
OCA Structure Browser,
and the
ConSurf Server
(which automatically colors amino acids by evolutionary conservation),
among others (see
Adoptions).
How Does It Work?
FirstGlance in Jmol is a user-interface to an excellent and free
molecular visualization program named Jmol
(more at
jmol.org). Jmol is very powerful, but in order to
access that power, you either need to learn a complicated command language,
or use a wrapper (user interface) such as FirstGlance in Jmol.
FirstGlance makes some of the power inherent in Jmol accessible to people
who don't have the time or inclination to learn Jmol's command language.
More..
What Can It Do?
FirstGlance in Jmol can display major structural features of the
molecule with one click each. Initially, each chain is given a
different color, making it easy to see how many protein or
nucleic acid chains are present.
Residues with incomplete sidechains are labeled S-,
and non-standard residues are labeled X automatically at the outset.
Clicking on something identifies it.
One-click options display secondary
structure, amino and carboxy (or 3' and 5') termini, composition
(protein, DNA, RNA, ligands, and solvent), the distributions of
hydrophobic, polar, and charged amino acids, salt bridges and
cation-pi orbital interactions for amino acids.
Non-standard residues and missing sidechains are flagged automatically.
Examples can be seen
in the
Snapshot Gallery.
A Find.. dialog makes it easy to locate amino acids by name,
sequence number, or a range of sequence numbers. Any chain, residue,
or atom can be hidden by clicking on it in the Hide.. dialog,
and remains hidden in most views until explicitly re-displayed.
The Contacts.. dialog shows the non-covalent interactions to any
moiety you select (by clicking on it). The non-covalent interactions
are automatically divided into seven categories:
hydrogen-bonded water, water bridges, hydrogen-bonded non-water, hydrophobic
interactions, salt bridges, cation-pi orbital interations, and metal and miscellaneous interactions.
Checkboxes control whether or not each category is displayed.
Examples can be seen
in the
Snapshot Gallery.
More Views.. includes coloring by uncertainty (B-factor, temperature
factor), and display of the unit cell, axes, or bound box.
Links to external resources show specific oligomers,
further information about the molecule and the quality of the model,
and locate any
gaps in the model.
What Should Beginners Use?
FirstGlance in Jmol was designed for readers of scientific
journals who are already familiar with the principles of protein structure.
Therefore, it lacks the extensive help, and guidance in
understanding, interpreting and evaluating 3D macromolecular
models that is built into
Proteopedia.Org
(see the list of articles in Proteopedia at
About Macromolecular Structure),
or the now-defunct
Protein Explorer.
We highly recommend
Proteopedia.Org
as the place to start exploring a macromolecular structure.
It offers a link to FirstGlance in Jmol from every
PDB code-titled page, as well as the publication abstract,
an evolutionary conservation scene,
and much more.
The
Proteopedia.Org
wiki is also the easiest place to create customized molecular scenes
(using its built-in Scene-Authoring Tools), and share them
immediately on Proteopeda web pages.
Ease of Use.
FirstGlance in Jmol was designed to enable viewing a molecule in one click,
in all popular web browsers and computer platforms -- without installing
anything. Several other java-based applications that meet these criteria
are offered by the
RCSB Protein Data
Bank. Generally, these lack the ease of use, automatically-displayed
context-sensitive help and color keys provided by FirstGlance in
Jmol, and offer a less complete set of one-click molecular overview options.
While FirstGlance in Jmol is designed to be "obvious" with labels
clear to non-specialists that avoid crystallographic jargon, the
other viewers often use the latter.
Touching controls in FirstGlance in Jmol raises explanatory "tool tips", absent
in the other viewers.
For those interested, most terms in FirstGlance in Jmol
are carefully defined (e.g. "metal", "carbohydrate",
"non-standard residue") while other viewers typically leave such
terms undefined.
Power.
None of the other viewers offer an easy method to show all
non-covalent interactions to any arbitrary moiety, comparable to
the Contacts.. dialog of FirstGlance in Jmol.
However, aside from the general limitations noted above,
each of the in-browser java-based viewers offered by RCSB has its special strengths:
QuickPDB has a nice
sequence-to-structure interface.
At the outset, QuickPDB shows a one-letter code sequence, and when you
touch (not click) any residue, it labels that residue in the 3D
view. FirstGlance in Jmol lacks a sequence listing at this time,
but you can use its "Find.." interface to enter a sequence number
or range to locate them in the 3D view. And FGiJ has a
PDB-code-specific link to RCSB for a sequence listing, and also a
"Gaps" link to the SEQRES to Coordinates (S2C) listing from Wang
and Dunbrack. However, QuickPDB (in late 2006) did not show nucleic
acids at all.
WebMol does dot surfaces including cavities (full molecule
dot surface rotates very slowly; FirstGlance in Jmol lacks cavity viewing). It
has very nice interactive distance matrix and Ramachandran
displays.
However, in order to run it, the user must approve an
invalid digital signature. The initial view is uninformative
wireframe (like RasMol). Touching a residue reports amino acid
but not nucleotide identities. It has a wire-thin backbone trace
which is not smoothed, and it offers no ribbons view.
KiNG (Kinemage New Generation) shines for specialized applications
such as showing clashes at the
MolProbity Server, but has many idiosyncracies that
limit its usefulness for general viewing by novices.
Image Quality.
FirstGlance in Jmol's image quality is better than any of the other
in-browser, java-based viewers offered at the Protein Data Bank,
even without toggling on the high-quality rendering mode in FirstGlance.
Nevertheless it does not provide
the "publication quality" images of PyMOL or the stand-alone Protein Workshop java
application (see below).
Protein Workshop, also offered by the RCSB PDB, is a stand-alone
java application spawned by the web browser (not an in-browser
view). When tested in late 2006, this was unacceptably slow to rotate
ordinary-sized proteins, but has much more beautiful rendering
than does FirstGlance in Jmol.
Viewers Not Offered by RCSB.
There are numerous, excellent stand-alone 3D molecular viewing
applications (most listed at The World Index of Molecular Visualization
Resources).
PyMOL
(
Overview,
PyMOL Website),
in particular, has become widely popular
among structural biologists in recent years, including for making
publication-quality images.
Where Do The Molecules Come From?
FirstGlance in Jmol usually obtains the molecules it displays
(atomic coordinate
PDB files) from the
US Branch
of the
Worldwide Protein Data Bank (PDB), which makes freely available
all published macromolecular 3D structure data. In order to display
functional biological assemblies (specific oligomers), theoretical
models, and other data files not available from the PDB, FirstGlance in Jmol
can obtain molecules from any web server
(More..), or you can upload your own model data.
How Does It Work? (Technical)
Technically, FirstGlance in Jmol is a user interface to the Jmol java applet.
FirstGlance in Jmol is written in javascript and HTML, while Jmol is written in
java.
As mentioned
above,
FirstGlance in Jmol makes the
visualization power built into the Jmol java applet accessible to
users who do not have the time to learn the Jmol command
language, or how to embed Jmol in web pages.
When you request a particular view in FirstGlance in Jmol, the Jmol commands
to produce that view are generated dynamically (with javascript) and sent
to Jmol. The user may be unaware of this, simply seeing the desired view.
Jmol is developed
separately by its own open-source volunteer team. FirstGlance in
Jmol has been developed largely by
Eric Martz
(author of Protein
Explorer), but others are free to get involved since it is an
open-source project. FirstGlance in Jmol uses the
Creative Commons Attribution-Noncommercial-Share
Alike 3.0 License,
while Jmol itself uses the
GNU Lesser General Public License.