Press Archive

Swami, the Next Generation Biology Workbench

Published 11/03/2005

By Lynne Friedmann for SDSC

Facilitating Science
An example of the kind of protein rendering that users can easily produce with the Workbench. To make the Next Generation Biology Workbench usable by the most researchers and students, the project is developing tools that run on machines at SDSC so that users don't have to download and install software on their machines. Credit: NGBW.

A free, Web-based interface that links major molecular biology databases with analysis programs-accessed thousands of times a week by scientists and students worldwide-is about to become even better.

In May, the San Diego Supercomputer Center (SDSC) at UC San Diego announced the award of $2.2 million from the National Institutes of Health (NIH) to build on the ground-breaking "Biology Workbench," introduced by pioneering bioinformatics researcher Shankar Subramaniam nearly a decade ago to provide broad access to many biology software tools and data resources through a Web-based, point-and-click computational environment. The Biology Workbench also offers the advantage of speed, letting researchers complete within hours work that once took days, weeks, or even months.

The Next Generation Biology Workbench (NGBW), also known as Swami, is a unique and versatile biological analysis environment that allows users to search up-to-date protein and nucleic acid sequence databases. Searching is integrated with access to a wide variety of analysis and modeling tools, all within a point-and-click interface that eliminates file format compatibility problems. This has made the Biology Workbench an indispensable resource for scientists who work in bioinformatics, an emerging discipline in which researchers collect, integrate, model, and analyze the explosion of biological data produced by such efforts as the Human Genome Project.

In addition to adding to fundamental scientific knowledge, this research can lead to improved understanding of disease and open the door to development of new treatments and drug discovery. The NGBW prototype, as well as links to the current Biology Workbench can be found at

"I have been using Biology Workbench on a regular basis for the last three to four years," said Vanderbilt University assistant professor, Mark de Caestecker. "It has proved to be an invaluable tool for the analysis and design of gene and protein constructs used in a range of different experiments in my laboratory. The Biology Workbench has the most comprehensive and easy-to-use applications I have come across." In his research, de Caestecker studies stem cell differentiation in kidney development, cancer, and tissue injury repair, and also researches cellular signaling in relation to hypertension.

Updating a Favorite

Easy Visualization
The NGBW will feature lightweight server side tools for visualizing and manipulating protein molecules, such as the scalable vector graphics viewer shown here. Credit: NGWB.

The researchers are making the Next Generation Biology Workbench even more useful by expanding its present offering of 65 tools. Like the original, the NGBW will continue as a free Web resource that offers access to data, data storage, software tools, and computational resources that help researchers mine the information in many popular protein and nucleic acid sequence databases. NIH funding will support the construction of up-to-date features such as improved user interfaces and an expandable architecture that will allow the NGBW to continue to evolve in the future in response to new developments in technology, biology, and the needs of scientists.

"There have been huge leaps in the technologies used in building cyberinfrastructure since the original Biology Workbench was created," said Mark A. Miller, SDSC project leader for the new grant. Work will be done in phases with a beta release planned for April 2006. According to Miller, there are some upgrades that can be accomplished in a matter of months, while others will take a year or two to accomplish. For example, the current workbench integrates information from 33 public databases, which are downloaded into a flat file. Using the less powerful technology of the flat file format places significant limitations on search functionality. Therefore, a major goal of the Next Generation project is to adopt a relational database format in which the information is broken down into tables and categories, which then allows more complex queries, or scientific questions, to be answered.

"Software developers always want to make things very elegant so they can later expand and make them more modular," said Miller. "We do want that, but we don't want to make people wait five years for the next product. So our focus is giving users something today and then making it more elegant underneath."

Other improvements will include enhanced visualization and data management capabilities, and to make sure that these services are available even to users with only a lower speed dial-up modem. This will enable a wide range of users to pose sophisticated questions, even if they don't have access to advanced computing resources.

While not losing sight of the researcher with limited financial resources for whom the Biology Workbench was originally developed, enhancements to the Next Generation Biology Workbench are expected to also capture more high-end users. The current workbench, running on a Sun computer, allows access to more than 32,000 active users, the majority of whom look at single sequences, submitting more than 120,000 requests for analysis, or jobs, monthly. This is not high demand, and consequently not very expensive computationally.

"We could probably handle four times that amount without breathing hard," said Miller. "But the current system is limited because of how the file system is structured. I believe we can design the Next Generation Biology Workbench so it will be fully expandable in the future."

A Tool for Education

Using the Biology Workbench in the Bioinformatics Teaching Lab at the University of Idaho. Credit: Michael Placke.

Because navigating the interface is comparable to learning the Windows or Macintosh operating systems, it didn't take long for instructors to embrace the original Biology Workbench as a teaching tool. Responding to this growing user segment, SDSC researchers are partnering with colleagues at the National Center for Supercomputing Applications (NCSA), where the workbench was initially developed, in developing an educational component. And consulting with educators in the early stages of the Next Generation Biology Workbench design, Miller explains, "will keep us from getting too far off the beam making a nice architecture but the wrong functions." The outcome will be a dedicated component or view for students and teachers, called the Student Biology Workbench.

This is welcome news to instructors like Celeste Brown, Bioinformatics Coordinator, Initiative for Bioinformatics and Evolutionary Studies at the University of Idaho. "Ten years ago a graduate student found the Biology Workbench (on the Web) and brought it to my attention," she said. "I've been using it for teaching ever since."

Brown is not alone. A Web search reveals a wide range of lesson plans designed specifically with the Biology Workbench in mind. Many are from smaller institutions of higher education, such as the University of Idaho. The use of the Biology Workbench in this setting should be welcome news to the NIH, which supports an initiative to encourage bioinformatics training in states that have historically not received as high a level of government grant dollars as, say, the state of California. In this program, 23 states and Puerto Rico qualify for additional NIH support for faculty development and enhancement of research infrastructure under the Institutional Development Award (IDeA) Program. While there isn't a requirement that IDeA states use the Biology Workbench to train students in bioinformatics, many do.

Under the Hood
A core goal of the project is to improve the Workbench using new technologies. The design goals include using an architecture that allows the NGBW to be freely available for distribution, and developed within the available budget and a short time frame. The team has addressed these issues by leveraging the Java Enterprise Edition software stack as implemented by the open-source JBoss 4.0 Application Server. The new workbench stores and retrieves data from relational databases by mapping Java objects to relational entities using JBoss' Hibernate persistence library. The user works with this data through a user-friendly, Web front-end that is implemented using the Apache Struts web-application framework.

"Because many users do not have authorization or sometimes the ability to install programs on their computers, we're developing very lightweight visualization tools that run effectively from the server side at SDSC," said Miller. This ensures that a wider range of users can benefit from NGBW data and tools.

Another challenge is how to support the wide variety of analytical tools made available within the Workbench, since such tools typically have very specific input and output format requirements. To do this, Miller explains, the developers must "wrap" each separate program, putting a translator between it and the central NGBW architecture, so that data supplied by the user can be passed to any of the analytical programs in the Workbench in the language it can interpret. In turn, the translator returns output to the user in a common format. And since the developers are using a well-defined common language, this also makes it straightforward for other developers to create their own tools to be added to or work with the Workbench.

At the University of Idaho, an evolutionary perspective is the emphasis of all biology training. "But let's face it. Nobody really likes development lab where you put an egg into a Petri dish and watch how it develops into a chicken," notes Brown. "A tool like the Biology Workbench puts things in context and reinforces scientific principles learned in earlier classes. Besides, students are used to a lot more technology than they were ten years ago."

Brown uses the Biology Workbench in an introductory enzyme lab course to access 3-D structures from the database. Students not only see what's going on in the reactions they set up, they can then consider how the structures evolved. "I want students to understand that there are databases out there that have all this nucleotide and protein sequence information," said Brown. "I also want them to realize that it's easy to get to and there are a lot of tools out there to help them analyze what's in those databases."

Outside the classroom researchers have become aware of the Biology Workbench through scientific publications. In many cases, at the end of a commercial software review, the Biology Workbench is mentioned as a free solution that scientists might also wish to consider. When the Next Generation Biology Workbench is ready for release, there is travel support in the NIH grant for a "roll out" of its new capabilities at a series of major national scientific meetings.

"The original Biology Workbench created a new paradigm for integrating biological information and tools, giving easy access to researchers as well as students," said Shankar Subramaniam, Professor of Bioengineering and Director of the Bioinformatics Program at UCSD. "It's rewarding to see the growing interest in this resource, and bringing the workbench forward using modern technologies will make this important tool more versatile and available to an even broader range of users."

When Miller recently posted an Internet request for testimonials about the Biology Workbench he soon heard from students, teachers, and researchers from the four corners of the earth. Feedback includes such superlatives as "invaluable," "easy to use," "faster and more efficient than other tools," and "critical for completion of my thesis research."

Professor of Biology Darrel Stafford at the University of North Carolina, Chapel Hill wrote, "We used it extensively in our search for the gene for epoxide reductase which was published in the March 5th issue of Nature."

As far as Gabriel M. Belfort, an MD/PhD candidate at Boston University School of Medicine, is concerned, "not having the Biology Workbench would be the functional equivalent of replacing my computer with an abacus."

The Next Generation Biology Workbench team at SDSC includes Mark Miller, PhD, PI; Mike Cleary, PhD, co-PI and user advocate; Shankar Subramaniam, PhD, co-PI; Kevin Fowler, senior software architect; Roger Unwin, database engineer; Gregory Quinn, PhD, senior interface engineer; and Ashton Taylor, artist. Celeste Brown, PhD, of the University of Idaho, is education advisor and bioinformatics coordinator.

-Lynne Friedmann is a freelance science writer living in Solana Beach, California.

Mike Cleary, Kevin Fowler, Mark Miller, Gregory Quinn, Ashton Taylor, and Roger Unwin, SDSC/UCSD
Shankar Subramaniam, UCSD and SDSC/UCSD
Celeste Brown, U. Idaho


Subramaniam, S. The Biology Workbench-a seamless database and analysis environment for the biologist. Proteins, 32(1):1-2, 1998.