Ten years ago, a student working in David Walt's chemistry lab at Tufts University conducted a frustrating experiment to create an array of pointed optical fibers for use as a new surface scanning method. Attempting to etch an optical imaging fiber array into these points, the student produced precisely the opposite result: the fibers etched into small wells or depressions in that array. Many mentors would have simply scratched the experiment and told the student to start over. Instead, Walt decided to try using the oddly-shaped array–and in the decade since, it has become a valuable technology used to analyze genetic profiles, screen drug compounds, detect chemicals, and more.
This experience exemplifies one of Walt's central credos: embrace the unexpected. "I generally tell all my students who do research to have a prepared mind and to realize that the most interesting experiments often emerge from something unexpected," said Walt.
Now, undergraduates will also have the opportunity to put Walt's philosophy into action as they work on open-ended projects that may produce unexpected findings and take them in new directions.
As an HHMI professor, the chemist plans to guide undergraduates in developing multi-year bioinformatics research projects. Students in the program will also design experiments for K-12 students and take a remodeled organic chemistry course, enhanced with chemical biology experiments. Walt's goal is to share contemporary, engaging science with students who ordinarily wouldn't encounter it.
Ultimately, Walt would like to fully integrate the undergraduate research experience, classwork, and outreach efforts. He wants younger students as well as undergraduates to know how dramatically the very pursuit of science has changed. Before genomics, proteomics, and other fields with huge data sets emerged, scientists all shared a traditional hypothesis-driven approach to investigation. Today, however, data often run the show, as researchers collect tremendous amounts of it and then use computer software to reveal complex patterns, which help them form conclusions and new questions.
"This model is a very different way to do science," he said. It mandates recruiting computer science students into the life sciences, breaking cultural barriers between scientific disciplines, and building teams." That's why he sees bioinformatics as such a good tool for teaching the new science.
"When I have undergraduates working in my lab, they inevitably say, `Wow, I had no idea all this incredible stuff was going on,'" Walt remarked. Gone are the days in which the primo class experiment involved squinting through a microscope at a cell scraped from the inside of a student's cheek, he said. "Today, if we are studying a cheek cell, we could do a complete genetic profile of that cell and tell the student his or her propensity toward certain diseases. This powerful capability is rarely conveyed to budding undergraduate science students. And I want to change that, not just for the students already in my lab, but for undergraduates generally."
