Long-Distance Laboratories

Collaboratories are changing the way scientists work with and educators teach about instrumentation.

One thing about Nestor, he's relaxed. It isn't often that, during an interview, the source props his feet on the table and leans back in his chair. You get the sense that Zaluzec, who is one of the leaders of two of Argonne National Laboratory's remote collaboration projects, isn't worried about the ultimate success of the programs. Give it time . . . we'll get it done, his body language seems to say.

Seeing is believing. Zaluzec at work.

Getting all of this input from just a telephone conversation seems a good trick, until you realize that Zaluzec has set up a remote control Web camera near his computer console. To check in on him, I just sent my browser to http://tpm.amc.anl.gov.

The explosive growth of the Internet over the past decade has Zaluzec and other forward-thinking scientists eager to incorporate the new technology into scientific research, but besides much speculation and some data-sharing sites such as GenBank, the chemical and biological sciences have yet to be transformed. Zaluzec and his colleagues, however, want to change that. They envision a world with increased access to scientific tools, expertise, and resources, where scientists communicate and collaborate easily with fellow scientists around the globe. Such Internet-bolstered partnerships are called collaboratories, and they seem to be gaining momentum (see Analytical Chemistry, 1997, 69, 741 A-42 A).

Instrumentation to the forefront

Although all manner of collaboratories are under way across different disciplines, the greatest amount of work has gone into remote access to instrumentation. It's a logical choice. Multimillion dollar price tags make top-of-the-line instruments inaccessible to researchers who aren't at big-gun government or corporate laboratories. Educators at small colleges and universities may have no access to certain types of instruments. Many of the lessons learned by sharing instrumentation can be applied to other kinds of collaborations.

Zaluzec's team has been at it since 1994. Zaluzec is the project leader of the telepresence microscopy collaboratory, which (through the materials microcharacterization collaboratory established by the Department of Energy in 1997) connects microscopes and their users at Argonne National Laboratory (ANL), Lawrence Berkeley National Laboratory (LBNL), the National Institute of Standards and Technology (NIST), Oak Ridge National Laboratory, and the University of Illinois-Urbana-Champaign. Researchers from all of the labs, as well as from industry and academia, can use instruments at these facilities from their own location.

"The whole point is not [just] to put remote instrumentation online . . . that's almost trivial. Collaboration occurs when we share resources, [which include] expertise, instrumentation, knowledge, and data," says Zaluzec. "You may want to collaborate with me not because I have a specific piece of instrumentation, but because I have expertise that can help with an instrument that you're working with."

After downloading the CORE2000 software (http://www.emsl.pnl.gov:2080/docs/collab/), users can collaborate using a variety of tools, including audio and video conferencing; multiple electronic "whiteboards"--the virtual equivalent to those ubiquitous whiteboards where scientists gather to sketch ideas and share data--which allow several users to see what one participant is drawing or pointing at; chat rooms, where users can have informal, "real-time" discussions; a televiewer that allows users to share portions of their computer screen with others; and an electronic notebook system, still under development, which accepts files and graphics, and can even accept data automatically downloaded from the instrument being used.

Remote NMR

The Environmental Molecular Sciences Laboratory (EMSL) at Pacific Northwest National Laboratory (PNNL) currently has projects under way in NMR spectroscopy and mass spectrometry, with remote use of other kinds of instruments planned for the future. In 1997, Jeffrey Pelton, an NMR spectroscopist at LBNL, began to use the 800-MHz NMR facility at EMSL through the collaboratory program.

"Live" microscope image of a precipitate (left) and an X-ray fluorescence spectrum showing that the precipitate contains Hg (right).

Pelton was working to determine the structure of a domain of a heat shock protein that is known to bind to DNA and influence transcription. The 500- and 600-MHz NMRs at LBNL wouldn't cut it--only EMSL's 800-MHz instrument had enough power to resolve the structure unambiguously. And because the biochemistry of the protein is well-defined, and it behaves well in solution, Pelton decided it would be a good problem to test as a collaboratory project.

Pelton lyophilized the material and sent it to Kelly Keating, an NMR spectroscopist at EMSL, who then prepared the sample for the NMR. After agreeing to a mutual time to log onto the system, each loaded theCORE2000 program, and Pelton logged onto the spectrometer and operated it remotely. Meanwhile, CORE2000's televiewer allowed Keating to see the screen in real-time and offer suggestions and guidance while Pelton manipulated the instrument.

Although using the instrument remotely has allowed Pelton to carry out the experiment without the expense and trouble of repeatedly traveling to the facility, he cites the importance of having met Keating in person before they began the experiments. "I haven't actually seen the spectrometer, but I've met Kelly. You need to trust each other to some extent." For example, if there's a precipitate in the NMR tube or the pH is not right, then Pelton has lost control of sample handling and the experiment could go awry. "You have to communicate and be more patient than you would in working with a spectroscopist whose office is just down the hall."

The collaboration was successful. Pelton assigned the peaks on the spectra for the region of the protein he's studying and plans further experiments on the instrument this month that should show the protein's three-dimensional structure.

Pelton sees a bright future in collaboratories, at least for the NMR community. "You have to have foundation money or institution money as well as NIH [National Institutes of Health] or NSF [National Science Foundation] money to purchase one of those [high field] instruments . . . the best instruments are so expensive that very few research groups can have one of their own."

Educating online

Researchers aren't the only ones with budget limitations. Science teachers at all levels have long been frustrated by lack of equipment. In addition, they must cope with textbooks describing long-outdated equipment while attempting to give students a taste of the real world of scientific research. Remote access to instrumentation and direct access to researchers through Internet technology could bring cutting-edge technology to the classroom.

Contents of an electronic notebook.

Zaluzec has been involved in just such a program. Last year's pilot effort involved Judith Bonkalski's sixth grade science class at St. Dominic's Elementary School (Bolingbrook, IL), which had just completed a unit on computers and their history, binary code, general electronics, and some simple circuitry. The grand finale was getting a close-up look at chips donated by Motorola and IBM under a 40x stereomicroscope.

Microscopists at NIST and Zaluzec at ANL set up views of similar chips on their high-powered electron microscopes. The students logged on through a broad-band telephone line and had direct control of the Argonne microscope, changing the angle and the magnification to peer layer by layer into the chip. "They didn't have to stop with the surface," recalls Bonkalski, now a principal at the school. "They knew how small the chip was, and yet when they looked at it under the electron microscope, they realized how something so small could be [very] complex . . . it helped them understand how the chip can perform all the [functions] that it does," she says. The lesson included three-way video and phone conferencing, with an IBM chip designer on hand--virtually--to answer questions about the chips.

The experience was a far cry from the traditional multiple-choice or fill-in-the-blank quiz at the end of a unit. "The neat thing about it is that as they learned more about the chip, they got access to equipment that would answer their questions," says Bonkalski. "It provided a real sense of closure."

The school plans to continue and expand the project. "It doesn't have to be one school and one scientist involved; we could have five schools and more than one scientist [online at the same time]. This could really increase the availability of expertise to schools," Bonkalski predicts.

Grade schoolers aren't the only ones getting the benefit of advancing technology. At Heritage College in Toppenish, WA, students in Hossein Divanfard's chemistry class are getting firsthand experience with PNNL's mass spectrometer.

Divanfard uses a large projection screen and audiovisual conferencing software to connect to a computer in the lab shared by John Price, a senior research scientist at PNNL, and Jeff Mack, a postdoc working under him. Mack then leads the class through a brief lecture based on materials that he has posted on the Web (http://198.28.64.33). As he moves from page to page, the browser on the class's big screen monitor moves right along with him. They move to a page with a Java application that allows the students to control the spectrometer remotely. Students can then run spectra of unknown samples and calculate distribution of isotopes and fragmentation patterns.

The virtual tour doesn't replace the classroom lecture; Divanfard starts out with an introductory lesson that lays the groundwork for what Mack will talk about. "Our job is to make [mass spectrometry] more concrete," says Price.

By all accounts, they've succeeded. Eight students from the class went on to do internships at EMSL the following summer. "The project is wonderful for small colleges like ours, because we will never be able to purchase expensive instruments . . . it's an excellent chance for students to gain experience," says Divanfard.

Price would also like to see the program expanded. "We'd like to present to more than one college at a time . . . and we'd like to take it up to the graduate level. [They could] develop their own master's or PhD projects around instruments that aren't necessarily at their own institutions. They could come out here and familiarize themselves with the instrument, then return and continue doing research," he says.

Online etiquette

The biggest barriers for collaboratories aren't likely to be technical, but rather sociological. E-mail text with no vocal inflections to accompany it can easily be misunderstood. Choppy audiovisual in the middle of a complex instrumental setup, or during a tense discussion of results, can leave a scientist staring at a frozen screen, wondering about an uninterpretable facial expression. The best way around those problems is to meet your collaborators face to face, says John Walsh, a sociologist at the University of Illinois at Chicago who has studied how scientists use the Internet. He cites research that suggests that "flames" and misunderstandings can be avoided "if the two people have met each other . . . in those instances when you receive e-mail and you're not quite sure how to interpret it, you're more likely to give your colleagues the benefit of the doubt if it's someone you've met."

Discussing NMR on a whiteboard.

Such discomfort extends beyond curious wording of e-mail messages. Thomas Finholt, director of the collaboratory for research on electronic work at the University of Michigan, cites the example of a group of neurologists and psychologists who are using the Internet to coordinate a search for genes that cause depression.

"The community had to come to some implicit understanding about data [posted on the Web] to protect participation by junior participants. The last thing a postdoc or junior professor wants is to have hard-won data poached," says Finholt. New recruits to the project are required to sign a covenant that spells out the rules.

Thankfully, collaboratories don't just create mistrust and trepidation. It offers advantages over traditional face-to-face communication, especially "in cases where you're coordinating with someone who doesn't speak English as [his or her] first language," says Finholt. "[These individuals] find it easier to follow a textual description rather than a spoken description. [For example], most Japanese students learn English in school, but the emphasis is on written expression and reading. They're at a systematic disadvantage when you're discussing things."

The future

Collaboratories will become more widespread as the tools continue to be refined, according to Raymond Bair, deputy director of the EMSL group that oversees and conducts much of the work on the institute's collaboratory projects. "In the past several years, we have spent a lot of time [developing a] critical mass of the more generic capabilities such as whiteboards, electronic notebooks, and other common applications," he says. "Now we're going to move into a more aggressive deployment and application phase. We've already started that in the NMR domain [with Pelton's heat shock protein collaboration] by building an NMR spectroscopist's notebook, creating tools to integrate molecular data into that notebook, and integrating them with the parameter files they need to manipulate the instruments. Each [scientific] domain has particular needs like that. If you can satisfy those needs, it completes the picture for that specialty."

Price sees one additional technical challenge: finding ways to integrate data from disciplines so that they can be "mined" on the Internet. "We have a 20-terabyte archive that we're filling out. That's a large problem in terms of managing and visualizing all of that information," he says. To tackle the problem, Price and his team are finding graphical ways to display groups of files and data. "Take mass spectrometry--what the spectrum gives you is chemical composition. Scientists are trying to relate chemical composition to the structure of proteins and DNA and other molecules of biological interest. Then, molecular structure has to be turned into an understanding of its function."

Now, take all of that information and multiply it by all of the disciplines that must come together to solve the problem--analytical chemistry, biochemistry, molecular biology, and others--and you begin to see the magnitude of the challenge. Does it seem daunting? Don't panic. Nestor Zaluzec certainly isn't. If you don't believe me, click on http://tpm.amc.anl.gov/ and see for yourself.

Jim Kling is a free-lance writer living in Bellingham, WA.