WebMO:
Web-based, state-of-the-art,
and cost effective computational chemistry
William F.
Polik (
Jordan R. Schmidt (
Computational Chemistry in
Education
Computer modeling has become widely
recognized as the third pillar of science, together with experiment and theory.1
Computation allows scientists to model realistic problems which do not have
analytical or simple solutions, and allows scientists to answer questions that
are difficult (either for technical or financial reasons) to address by
conventional experimental means. In chemistry, computational modeling is used
extensively in academic research and by the pharmaceutical industry. Computational
chemistry provides insight into chemical structure, interactions, and
reactivity, which in turn permits the prediction of chemical behavior. A high
fraction of chemistry research articles include some modeling to analyze and
interpret results, and modeling can be used to predict promising leads for
subsequent experimental investigation.
Some progress has been made to
introduce computation into the undergraduate chemistry curriculum, often as an
isolated laboratory experience or a specialized upper-level course. But computational
chemistry has not yet become infused throughout undergraduate chemistry
education, like chemical synthesis or spectroscopic characterization. Many of
the technical issues associated with scientific computing have been recently
overcome. For example, typical student laptop computers have more computing
power than supercomputers of 15 year ago2 and high-end research
computers of a few years ago. Graphical user interfaces and menu systems make
computers much more usable than before. Current computer codes3 can
accurately compute chemical energies to within a few kcal/mole and vibrational
frequencies to within 20 cm-1. A wide variety of exercises have been
developed for use in undergraduate chemistry courses.4
Yet despite the value of chemical
computation and the technical solutions to its application, three significant
barriers exist to the widespread use of computational chemistry in education:
WebMO fully solves these
ease-of-use, resource, and accessibility challenges. Molecules are drawn with a
3-D editor, selections are made from menu systems, calculations are run using
popular state-of-the-art computer codes, and results are viewed graphically or
in formatted tables. WebMO provides an easy-to-use, uniform interface to nearly
all popular computational chemistry packages. Only one server computer and one
license for commercial software is needed, dramatically reducing the hardware
and software costs. Since no software is installed on client computers,
time-consuming maintenance issues are also minimized. WebMO requires only a web
browser on any computer (Windows, Macintosh, Unix) connected to the internet,
making computational chemistry universally accessible to anyone from any
location. For example, students can do computational chemistry exercises from
their dorm rooms using their laptop computers.
WebMO Interface to Computational
Chemistry
WebMO
is a web-based interface for computational chemistry programs.10 WebMO
makes it possible to set up, run, and visualize state-of-the-art chemical
calculations from any computer using only a web browser. WebMO installs on a
single unix computer (Linux, Macintosh OS X, Solaris, Irix, etc.), after which
it is accessed and administered from a standard web browser from any computer
(Windows, Macintosh, Unix). And most importantly, WebMO is
free. A commercial add-on to the free package called WebMO Pro is
also available for advanced users.
After logging into WebMO, the user
can draw a molecule using a 3-D editor or
import a structure from a variety of formats. The molecular editor provides for
adding missing hydrogen atoms using organic chemistry rules, idealizing the
geometry using VSEPR rules, inserting molecular fragments, manually adjusting
the molecular geometry, and minimizing the molecular energy using molecular
mechanics. Creating a molecule by clicking to insert atoms and dragging to
insert bonds is extremely intuitive and requires very little student training. Advanced
users can specify the Z-matrix used to represent the molecular geometry, as
well as the internal coordinates to be fixed, optimized, and/or scanned.
After choosing one of the installed
computational engines, the calculation type
(Geometry Optimization, Vibrational Frequencies, Molecular Orbitals, etc.) is
selected from a dropdown menu. Common job options (Theory, Basis Set, Charge,
Multiplicity) may be specified, although reasonable defaults are provided. Additional
options (Symmetry, Solvent, Coordinate System,
|
|
WebMO 3-D editor |
Job options page |
When a job is submitted, it is sent
to the WebMO job queue where it is run in turn
at the first available opportunity. WebMO Pro also allows jobs to be run on a
remote computer or using the system's batch system (PBS, NQS, Sun Grid Engine).
While running, the job is continuously monitored and the partially complete raw
output can be examined.
When completed, the results can be visualized in 3-D, and bond lengths, bond angles, and
dihedral angles can be measured. Partial charges and the dipole moment are both
tabulated and displayed visually. Vibrational frequency calculations animate
the normal modes and display an infrared spectrum. UV-VIS and NMR spectra can
also be displayed. WebMO Pro also displays coordinate scans, molecular orbitals
and their dependent functions, and spreadsheet summaries that can compare
different calculations. Molecular geometries can also be exported into mol,
pdb, or xyz files for use by other applications.
|
|
Infrared spectrum |
Normal mode |
|
|
Molecular orbital |
Electrostatic potential |
Queued, running, and completed jobs
are managed using the Job Manager, which
includes a variety of features including uploading and downloading jobs,
setting user preferences, and user-defined folders for organizing jobs in WebMO
Pro.
A variety of administrative
tools are provided, such as setting of job time limits, creation of
student accounts from Excel lists, designation of groups of students that
correspond to different classes and/or different privilege levels, filtering
jobs by user or group, viewing jobs, and management of jobs. These features are
designed to permit a single instance of WebMO Pro to be used by multiple
faculty members who oversee different classes of students. The WebMO
administrator can also configure WebMO settings, check for updates, and configure
settings for individual computational programs using the web interface.
|
|
Job manager |
Spreadsheet summaries |
Since WebMO is web-based, no
software is installed on a user's computer. Only the web browser that comes
with every Windows, Macintosh, or Linux computer is needed to use WebMO. Thus,
prospective users can visit the WebMO Working Demo at
and try WebMO just as easily as visiting any other website.
Capabilities of WebMO
WebMO supports a variety of popular,
state-of-the-art computer programs:
and works fully with a variety of web browsers, including:
The specific computational jobs that
can be run depend on the capabilities of the underlying computational chemistry
program. Calculations that can be run and/or visualized include:
(* indicates WebMO Pro feature)
Cost of WebMO
Since WebMO, the linux operating
system, and some of the underlying computational engines are free, a very
capable computational chemistry system can be assembled at little to no cost. Should
one wish to license commercial software, purchase a WebMO Pro license, or
purchase state-of-the-art hardware, the cost will increase. Regardless, the
total cost of a WebMO system will typically be many times less than a computer
lab outfitted with computational chemistry software, and it will typically
serve many more students. The range of typical costs associated with WebMO is
shown below.
Item |
Cost |
PC (1GHz, 256MB, 40GB minimum) |
$ 0 - 1000 |
Linux OS |
$ 0 - 100 |
Computational engines (Gamess,
Gaussian, Molpro, Nwchem, Mopac, Qchem, Tinker) |
$ 0 - 2500 |
WebMO or WebMO Pro |
$ 0 - 995 |
TOTAL |
$ 0 - 4600 |
Conclusions
Computational chemistry is becoming
an increasingly important part of every chemist's training. Although today's
computer hardware and software are capable of high accuracy chemical
calculations, concerns about ease-of-use, time, money, and accessibility are
still preventing the widespread use of computational chemistry in the
undergraduate curriculum. WebMO is a web-based interface to computational
chemistry that solves these problems. Students are able to use a web browser on
their personal computers to easily setup and run state-of-the art calculations
using a graphical interface. WebMO installs on a single server computer,
greatly reducing hardware, software, and maintenance costs. WebMO can be
accessed from any computer on the internet, resulting in universal
accessibility. Thus, undergraduate students in any chemistry course, from
general chemistry through undergraduate research, can readily use WebMO to
perform computational chemistry calculations as part of their chemistry
education.
References
1. Report of the High-End Computing Revitalization Task
Force, "Federal Plan for High-End Computing,"
2. Top 500 Supercomputer Sites, http://www.top500.org/ (accessed June 2006).
3. F. Jensen, "Introduction to Computational Chemistry,"
John Wiley & Sons, 1999; C.J. Cramer, "Essentials of Computational
Chemistry: Theories and Models," John Wiley & Sons, 2002.
4. W.J. Hehre, L.D. Burke, A.J. Shusterman, W.J. Pietro, "Experiments
in Computaitonal Organic Chemistry," Wavefunction, Inc., 1993; J.B. Foresman
and AEleen Frisch, "Exploring Chemistry with Electronic Structure Methods,"
2nd Ed., Gaussian, Inc., 1996; M.L. Caffery, P.A. Dobosh, D.M. Richardson, "Laboratory
Exercises using HyperChem," Hypercube, Inc., 1998; W.F. Polik and J.R.
Schmidt, "WebMO User's Guide," WebMO LLC, 2003.
5. Gaussian.com, http://www.gaussian.com/
(accessed June 2006).
6. Wavefunction, Inc, http://www.wavefun.com/
(accessed June 2006).
7. Fujitsu Computer Systems BioScience Group Cache, http://www.computers.us.fujitsu.com/www/products_bioscience.shtml?products/bioscience/cache
(accessed June 2006)
8. Serena Software, http://www.serenasoft.com/
(accessed June 2006)
9. NWChem Home Page, http://www.emsl.pnl.gov/docs/nwchem/
(accessed June 2006); Ecce: Extensible Computational Chemistry Environment, http://ecce.emsl.pnl.gov/ (accessed June
2006)
10. WebMO - Computational Chemistry on the WWW, http://www.webmo.net/ (accessed June 2006).