June 2004
http://www.coe.berkeley.edu/labnotes/0604
Commuter-friendly version

Professor Adam Arkin was named one of Time magazine's "Time 100 Innovators" in the year 2000. (Peg Skorpinski photo)

If you could snap together genes, proteins, and cells like Tinkertoys to build complex systems that don't already exist in nature, what would you build? UC Berkeley researchers have no shortage of ideas. For example, a microbial "factory" that produces an antimalarial drug could cut the cost of pills from dollars to dimes, saving millions of lives every year. How about microrganisms that glow red in the presence of certain environmental contaminants and digest the toxins?

This is the vision of synthetic biology, building living systems from the bottom up to do our bidding. Bioengineering professor Adam Arkin is pioneering the computer modeling tools that will help make that vision a reality. The aim is to develop software for designing new living systems from a library of validated interoperable genetic "parts" with specific functions.

"Most genetic engineering is done by hook-or-by-crook," Arkin says. "It takes a lot of trial-and-error to build simple things into cells, like the ability to produce a lot of a functional protein. Now though, we want to actually program cells as if they're computers so they can do much more complicated tasks."

In some ways, synthetic biology is not as far removed from computer design as one might think. Genetic "circuits," comparable to electronic circuits, consist of specific sections of DNA that interact to produce a predictable result. As the specifications become more exacting, Arkin adds, the complexity of the bioengineering skyrockets.

Arkin's frequent collaborator, chemical engineering professor Jay Keasling, made a major biological circuit breakthrough last year. He engineered E.coli so that the bacteria produces the precursor to artemisinin, a chemical compound used to fight malaria. Currently, artemisinin is extracted from the leaves of the wormwood tree, an expensive process. Keasling transplanted genes from the wormwood tree into E.coli and assembled them into a new metabolic pathway enabling the bacteria to spew out artemisinin. Eventually, the same technique could be used to produce drugs that target cancer or HIV. While Arkin calls the work ‘stunning science', he points out that it took several years for the artemisinin factory to step up production.

"Jay didn't have the mathematical models to guide him or the biological parts in the freezer," says Arkin, who is also a faculty scientist at the Berkeley Lab's Physical Biosciences Division. "Our aim is to make the process he used orders of magnitude more efficient. So we're trying to develop a principled approach to creating a store of characterized genetic circuits and parts that we can then use much like the electronics industry uses transistors and capacitors. "

Continuing the metaphor, Arkin is pushing ahead on the development of Berkeley BioSPICE, software that represents and simulates cellular processes such as gene expression and cell division. Think of it as CAD (computer aided design) for genetic circuits. A project of the California Institute for Quantitative Biomedical Research (QB3), Berkeley BioSPICE analogous to SPICE (Simulation Program Integration Circuitry Evaluation), the industry-standard tool for integrated circuit design invented at UC Berkeley. The project's motto, "open source biology," refers to the fact that the fruits of the research are freely available for anyone to use and improve upon.

The idea is that once accurate computer models of the various cell behaviors and genetic circuits are created, researchers will be able to use BioSPICE to model and test new synthetic biosystems before they are fashioned in the wet lab.

"We can engineer these systems in theory," Arkin says. "And the multidisciplinary connection with Jay enables us to test out our ideas in the experimental world."

For example, Mike Cantor, a PhD student in Arkin's group, is working with Keasling to design a genetic "pulse generator," a device that responds to a biological signal and responds by generating a protein. Such a device could increase our understanding of genetic processes while helping characterize new parts. Meanwhile, former biophysics graduate student Leor Weinberger is leading an effort with Arkin and chemical engineering professor David Schaffer to develop early mathematical models for synthetic viruses that might someday be used to treat HIV.

"One of my mottos is that you don't really understand a system unless you can predict, control, and design in its media," Arkin says. "So right now, we're in the stage where we're determining if we can manufacture biological systems to spec."

Professor John Canny and graduate student David Nguyen (left) beside their MultiView system. (David Pescovitz photo)

Three-dimensional videophones are usually lumped into the category of science fiction, technology that's rooted more in The Jetsons than in reality. Not for much longer though. UC Berkeley researchers have developed a system that adds depth to today's teleconferencing software without requiring the user to wear bulky 3-D glasses.

Developed by computer science professor John Canny and PhD student David Nguyen, the InterView system enables participants to view images displayed on a large screen in three dimensions. An animated pendulum appears to swing out from the screen. In another demonstration, multiple viewers can study a collection of computer-generated spheres from various angles. As startling as these naked-eye effects are, they barely hint at what Canny and Nguyen envision for the project.

"Videoconferencing today isn't very good," Nguyen says. "When a 3-D scene is projected as a 2-D image, important non-verbal cues are lost. Our system supports mechanisms like gaze and body language that help you interact with specific people in a room."

InterView is one of the inaugural projects of the new Berkeley Institute of Design (BiD). Affiliated with the Center for Information Technology Research in the Interest of Society (CITRIS), BiD is a cross-disciplinary research center spanning human-computer interaction, architecture, product design, art practice, education, and engineering. InterView was born from an effort sponsored by the non-profit Corporation Education Network Initiatives in California (CENIC) to develop high-bandwidth applications for the next generation Internet.

Nguyen and Canny view a 3-D image of a swinging pendulum from opposite angles. The image is not clearly visible in this still photo. (David Pescovitz photo)

"CENIC wanted to identify why people would want gigabit networks in their home," Canny says. "3-D videogames and television seemed to us like plausible killer apps. However, we're starting with videoconferencing because it is a compelling commercial application and at a price point that is close to traditional corporate 2-D videconferencing solutions."

The InterView system consists of a bank of sub-$100 videocameras that capture the remote space. These images are streamed over the Internet and projected onto a custom display fashioned from inexpensive optical materials. The display carefully controls the multiple streams such that a viewer's left and right eyes can receive different images, resulting in the stereoscopic effect. Today, an InterView system might cost $50-$100,000 due to the high cost of the digital projectors involved. If 3-D conferencing takes off though, the researchers expect the hardware prices to drop low enough for their system to become practical for the home market.

"There are cost-effective solutions," Canny says. "For our system, each projector's display power can be much lower. Those kinds of little projectors could conceivably cost less than $100 each. And at least one company is developing products in that space."

For now, the researchers are honing their technology for commercial applications. One of the key problems of 2-D videoconferencing that they hope to solve is the "Mona Lisa Effect," the way that the eyes in Leonardo's painting seem to follow every onlooker all at once. This optical phenomenon makes it extremely difficult for a remote participant to establish eye contact with a single individual when multiple people are present. Similarly, an important mechanism of collaboration, "deixis," is lost in 2-D displays. Deixis refers to the use of gestures or gaze to bring someone's focus to a particular object.

"If I point at something in a remote space, that gesture gets warped by the translation into 2-D," Nguyen says.

InterView preserves gaze and deixis by making the image that any viewer sees dependent on their viewing angle. That way, Canny explains, when a remote participants gestures to something in the room on the other end of the teleconference, there's no question what he's pointing at.

Currently, Canny and Nguyen are modifying Microsoft ConferenceXP, an open research platform of software for collaborative online applications, to support the InterView capabilities. This summer, they plan to conduct a complete two-way 3-D videoconference.

"We've spent many years developing systems like robots and other technologies to make remote presence much more realistic," Canny says. "InterView is the culmination of that work."

UC Berkeley Civil and Environmental Engineering professor William Nazaroff.

Everyone knows that secondhand smoke stinks. But what is the real impact on the eleven percent of the U.S. population, the 31 million Americans living with people who puff in the house? UC Berkeley Civil and Environmental Engineering professor William Nazaroff and Brett Singer, a research scientist at Lawrence Berkeley National Laboratory (LBL), are studying the science of cigarette smoke to better grasp the scale of the problem. Careful analysis of air contaminants could help identify ways to reduce the health risks for people whose housemates won't kick the habit.

"Environmental tobacco smoke is more than a black and white problem," Nazaroff says. "Previously, it's been studied as an entity rather than subdividing it into specific toxic components."

For their study, the researchers calculated the amount of certain toxicants emitted in homes that are attributable to smoking and inhaled by the nonsmokers living there. They then estimated the amount of those same toxicants emitted by industry into outdoor air and breathed by everyone in the United States.

"The irony is that the amount of several of these pollutants breathed by those individuals exposed to environmental tobacco smoke far outweighs the total amount that all the rest of us breathe by living in polluted urban air," he explains.

In May, the Nature Publishing Group highlighted the results of the researcher's latest study, published in the Journal of Exposure Analysis and Environmental Epidemiology. The researchers analyzed 16 hazardous air pollutants including benzene, acetaldehyde, and formaldehyde, chemical species that are known or suspected carcinogens or linked to other serious health effects such as reproductive problems and birth defects.

The "smoking room" at Berkeley Lab features a machine that puffs away and sensors to monitor the pollutants in the air. (courtesy the researcher)

Nazaroff and Singer explore the physical science of smoldering cigarettes in a laboratory appointed as a smoking lounge. "It's a good thing we have a machine doing the smoking and not a human," Nazaroff says.

The room is instrumented with sensors to measure the change in pollutant concentrations.

"The experiments help us understand indirect smoke exposure, the chemical characterization of the emissions from tobacco smoke, and ultimately how much pollution comes from each cigarette," he explains

Meanwhile, the researchers are also studying data gathered by Nazaroff's former graduate student, Neil Kepeis, about indirect exposure to tobacco smoke in homes and how things like portable filters, closed doors, and open windows, affect air quality. The aim is to develop empirically-proven guidelines that could help non-smokers reduce health risks.

Molecular Foundry ceremonial groundbreaking, left to right: US Representative Mike Honda; Molecular Foundry director and professor Paul Alivisatos; Patricia Dehmer, DOE Associate Director of Science for Basic Energy Sciences; Sean Randolph, president of the Bay Area Economic Forum; Berkeley Lab director Charles Shank. (courtesy LBL)

When it opens in 2006 on the hill overlooking UC Berkeley, the Lawrence Berkeley National Laboratory's Molecular Foundry will be the crown jewel of the University's pioneering efforts in nanoscale science and engineering. On January 30, the facility's director and chemistry professor Paul Alivisatos--who holds also holds a faculty position in the department of Materials and Science Engineering--picked up a shovel for the official groundbreaking ceremony marking the start of construction on the $85 million, six-story research facility.

"Right now, scientists are inhibited in what they can do by the fact that they can't always get access to the materials they'd like to work with," Alivisatos says. "The Foundry will keep track of what nanofabrication techniques work well and make those available so all scientists can become experts. That will certainly accelerate the pace of innovation."

The SmithGroup of San Francisco designed the Molecular Foundry building. (courtesy LBL)

The Foundry is one of five Department of Energy Nanoscale Science Research Centers to be constructed in the next few years. The aim is to establish a hub for collaboration between researchers from such diverse disciplines as materials science, biology, electrical engineering, physics, and chemistry. The novel devices expected to emerge from the foundry range from incredibly precise nanosensors for the detection of environmental contaminants to highly-efficient and inexpensive flexible solar cells to ultra-fast nanocomputers.

"Berkeley is blessed with tremendous resources, such as the national supercomputing center (NERSC), the Advanced Light Source, and the National Center for Electron Microscopy," says US Secretary of Energy Abraham Spencer. "All will be instrumental in the revolution in science offered by the Molecular Foundry."