 |
|
Analyze this
Bioinformatics analysis is shaping the future of medicine
By Darv Johnson
Imagine that you are given a simple task: to distinguish blue objects from yellow ones. The job is so easy that you can perform it by hand. Now imagine that, instead of two colors to distinguish, you suddenly must sort 100 colors, some of which look very much alike.
That's the analogy Dr. Stephen Winters-Hilt, an assistant professor of computer science at the University of New Orleans with a joint appointment at the Research Institute for Children at Children's Hospital New Orleans, uses to describe the dilemma facing today's medical researchers. Scientists have become so good at generating data via sophisticated technologies, he says, that they are drowning in it. "It's bad enough now and a year from now, it's going to be a whole lot worse," Winters-Hilt says of the data deluge. "It's a ready-made problem for someone on the computational side."
Winters-Hilt, whose second Ph.D. is in computer and information science (his first is in theoretical physics), has the know-how to help turn mountains of data into medical solutions. He is an expert in the exploding field of bioinformatics, which promises to shape the future of medicine and health care.
In its broadest definition, bioinformatics is the use of computers to process and analyze biological information. The widely publicized Human Genome Project, for example, which created a map of the entire DNA sequence of the human body, is perhaps the most famous bioinformatics achievement to date. "You're studying the flow of information in living systems," says Winters-Hilt.
His colleague Seth Pincus, M.D., director of the Research Institute for Children at Children's Hospital New Orleans, calls bioinformatics a "key element of modern biology."
"We are now able in biology to generate huge amounts of data, so much data that one human alone can't interpret it," Pincus says. By way of explanation, Pincus describes the challenge faced by a scientist studying the HIV virus. If the scientists are using microarrays- samples of many genes arranged in a regular pattern-to examine every gene in the human genome, looking for the ones that are turned on during HIV infection, they are instantly faced with a deluge of information. A single sample alone, Pincus says, might include 40,000 data points.
"We are talking about millions of data points we need to analyze," Pincus says. "There is commercial software available, but it takes a lot of time and effort to use, and it turns out it is flawed and not entirely useful." But with Winters-Hilt, Pincus now has an in-house advantage. Drawing heavily on a branch of artificial intelligence called machine learning, Winters-Hilt can write tailor-made software to extract the data that is relevant to the problem at hand.
Bioinformatics techniques can be applied to a range of problems, from run-of-the mill medical endeavours like reading X-rays to the oft-discussed pursuit of DNA sequencing, which in its current incarnation requires loads of time, cash and computer power. The bioinformatics goal there, Winters-Hilt says, is to drive down the cost of sequencing an individual's genetic information from $10 million-roughly what it would cost now-to $1,000. At that cost, he foresees a day when people could have their relevant genes sequenced as a matter of course. Doctors could then analyze the data in search of genes that might cause trouble several decades down the road-a cancer-causing gene, for example, that isn't activated until middle age.
"In the future, doctors are going to look at (the information produced by bioinformatics analysis) and give you 15 years' advance warning that you have a propensity for a certain type of cancer," Winters-Hilt says, noting that bioinformatics is already part of the core curriculum for medical students. Thus forewarned, the doctors could then take steps to head the disease off before it even arrived.
Winters-Hilt compares the potential impact of this technology to that made by antibiotics, which transformed medicine and lengthened life expectancies with their widespread introduction in the 1940s. "It's going to revolutionize medicine," Winters-Hilt says. "I can't say it more strongly than that."
As for how long it will take to knock $9.99 million off the price of sequencing, he predicts sooner rather than later. "We're going to try and do it in the next five years," he says. "The race is on."
Who will win it? UNO's bioinformatics program, though small, is growing by leaps and bounds. Winters-Hilt has four students from the University of New Orleans computer science program working with him on an ongoing basis, and teaches bioinformatics in the classroom to another 15 students. At the same time, Winters-Hilt and other colleagues in the computer science and mathematics departments at UNO meet about once a week to discuss the latest challenges in their field. Winters-Hilt calls UNO's work in bioinformatics "regionally competitive"; that is, on par with that of the big research institutions in Texas and at Washington University in St. Louis. His goal is national competitiveness-the realm of Harvard, MIT and Stanford.
There's plenty of other competition. Big companies like Agilent Technologies, a Hewlett-Packard spinoff, are also directing millions of dollars into bioinformatics research and DNA sequencing. The 38-year-old Winters-Hilt, who has already been involved in several start-up ventures, knows the for-profit model has its benefits. That's why he also wants to bring his entrepreneurial acumen to bear on New Orleans and use it to build up the local biotech industry.
"Regardless of how we eventually crack this problem," he says, "We are definitely going to crack this problem."
|
|
|
|
