From biomedical engineering and medical device development to genetic research
and biological process technology, IT researchers play a major role in the growing
field of biotechnology. Here is a look at 13 of the most innovative projects
underway.
Genetic engineering promises to revolutionize both medicine and
agriculture by giving doctors new tools for diagnosis and treatment
and by enabling biologists to create new strains of fruits, vegetables,
and grains that are hardier, more nutritious, and more flavorful.
But those breakthroughs won't happen overnight. Before researchers
can precisely manipulate an organism's DNA to create a desired effect,
they must map and analyze its entire genetic code, or genome, to
determine which genes correspond with specific traits. That process
involves collecting and organizing enormous amounts of genetic information.
"Researchers are generating [genome] data on a scale no one
ever dreamed of before,” says Ernest Retzel, director of biocomputing
for the University's Academic Health Centers. The challenge, he
says, is to find a way to organize and analyze unprecedented amounts
of data.
Toward that end, he and an interdisciplinary research team led
by Professor Vipin Kumar are designing new computer techniques for
working with genetic data.
"At one time, data acquisition was the slow part,” says
Retzel. “We could analyze data using old computer techniques
on new CPUs and keep ahead. Now, [researchers] are generating so
much data, the bottleneck is in the analysis."
Analyzing genetic data is unusually difficult because so many variables
in the data are interconnected, and different labs have varying
standards of data collection. Moreover, few existing techniques
apply to this kind of “high-dimensional” data.
"This is perhaps the most challenging data-mining problem
that exists,” says Jaideep Srivastava, an associate professor
of computer science and member of the research team. “The knowledge
to be gained is of the most complex kind."
To meet that challenge, the team is developing new computer programs
to help biologists integrate and visualize genetic data from various
sources and examine it to identify structural and functional patterns
linked to specific traits. Once researchers link patterns to functionality,
they can begin to manipulate both.
"You don't need to think of a cell as magical anymore,”
says Retzel. “It's a system you can tweak like anything else."
Retzel and research associate Elizabeth Shoop provide the biological
expertise and data around which the project is built. Kumar, Srivastava,
and fellow computer scientists Ravi Janardan, George Karypis, and
Shashi Shekhar are working to formulate and solve the computational
problems. Once those problems are solved, the team will produce
a collection of data-mining, data-integration, and data-visualization
software for other biologists to use.
The project is being developed and tested using genetic information
from a wide variety of plants—including soybeans, rice, eucalyptus,
and corn—because the group wanted to avoid the ethical issues
that might arise from working with animal or human DNA. They also
share a common concern about the world's food supply.
"Food production and population growth are way out of sync,”
says Retzel. “The 21st-century problem will be sustainable
agriculture.” He hopes the team's research will speed development
of genetically enhanced crops to better meet the nutritional needs
of the world's burgeoning population.
"For example, 90 percent of the world relies on rice as a
major source of food,” he says. “Using these tools to
enhance its yield and improve its resistance to disease should be
a top priority.” The group is collaborating with researchers
in other University departments and in universities and companies
around the world to achieve that goal.
Unfortunately, creating successful new breeds of super-crops may
prove far more complicated than mixing and matching genes. “If
you change the data in a tomato to make it stay ripe longer, how
will that affect its nutritional value?” asks Kumar. “We
don't know."
But the pace of genetic research is accelerating, and computer
tools may someday allow biologists to predict those effects. “The
techniques we're developing will allow molecular biologists to ask
and answer questions that they cannot even consider asking today,”
says Kumar.
