Wiggins Featured in Columbia News and The Record

Columbia News

Would you believe that the data mining technology that helps presidential candidates tailor their messages and helps grocery stores decide on sale items could also lead to breakthroughs in cancer research?

Chris Wiggins, assistant professor of applied mathematics, and his Columbia colleagues think that applying targeted algorithms to historical information to predict patterns of future behavior -- data mining -- will have a major impact on the understanding of many human diseases, including cancer. To this end, The National Institutes of Health has awarded Columbia University a five-year, $18.5 million grant to establish a National Center for Biomedical Computing. The center, to be known as the National Center for Multi-Scale Analysis of Genetic and Cellular Networks (MAGNet), will be housed at the Columbia University Medical Center campus in the newly established Center for Computational Biology and Bioinformatics (C 2B 2).

"Computational biology, a new field of science, has the potential to revolutionize biology and the translation of biology into medicine," said Andrea Califano, MAGNet director and professor of biomedical informatics.

"The goal of the National Centers for Biomedical Computing is to make it easier for the wider scientific community to exploit the power of computers to address fundamental biological and biomedical challenges," said Califano.

"We're trying to build new mathematical and computational approaches that would be directly applicable to medical advances," Wiggins said.

In the last decade, there have been a number of advances in new technology for large-scale data-driven biology, including biotechnology and data mining. And in the last five years, significant advancements have been made in analysis of real-world networks. The intersection of these advances has resulted in the new center.

"Machine learning, or data mining, is about making predictions," Wiggins said. "The challenge for people who are trained in natural sciences, like me, or a biologist, is to make models that are not only predictive but interpretable, because most biologists don't care how well you can predict how an individual gene will respond. The biologist is interested in knowing why it works that way: which parts of the genome are talking to each other -- the parts of the genome where edges exist in the genetic regulatory network that they didn't know before."

Wiggins is also part of a team that received a $3.7 million grant from the National Institutes of Health for a " Nano-Medicine Center for Mechanical Biology."

"Nano-medicine is not an established field yet," Wiggins said. "Our proposal is really an approach to build nano-scale devices to look at cellular biophysics. The particular problem we are examining is how force is detected and generated. When you push on a cell, how does a cell know, and how does physical information become chemical information, and how does chemical information become genetic information?

"We've already made progress quantifying what I'd call the phenotypic response of the cell to different chemical changes. For example, if I put a drug in a cell it actually moves differently and you can quantify that and see the time scale, how it steps, or the symmetry with which it spreads. Although you see patterns by eye, you really need the data to know if those patterns are something that you are imposing on the data. You need to let the data speak for themselves. We want to take these images and turn them into numbers that will then be interpretable to the biologists. So you turn these numbers into biology and then guide and constrain microscopic modeling, like we are familiar with in physics and applied mathematics."

Wiggins credits information-sharing through the Internet for advancing the field of computational biology, both by allowing researchers around the world to analyze data and by promoting biotechnical, mathematical and computational advances in machine learning.

"I'm very excited about the interdepartmental and interdisciplinary collaborations I'm working on here at Columbia, and the recognition they are now getting," he said. "Each of these grants is a very rare and special award that tells the research community that Columbia is a leader in biological computation."

 


Math Professor Poised to Make Fresh Strides in Computational Biology
The Record, Volume 31, Issue 5, p. 5
November 14, 2005

Data mining technology, which applies targeted algorithms to historical information to predict patterns of future behavior — and helps presidential candidates tailor their messages to voters and grocery stores decide which items to put on sale — may also lead to breakthroughs in cancer research, in the view of assistant professor of applied mathematics Chris Wiggins.

Wiggins is part of a multi- disciplinary team that recently received a five-year, $18.5 million grant from the National Institutes of Health to establish a National Center for Biomedical Computing, to be known as the National Center for Multi-Scale Analysis of Genetic and Cellular Networks (MAGNet).

“The goal of the National Center for Biomedical Computing is to make it easier for the wider scientific community to exploit the power of computers to address funda- mental biological and biomedical challenges,” said MAGNet director Andrea Califano, professor of biomedical informatics.

“Machine learning, or data mining, is about making predictions,” explained Wiggins. “The challenge for people who are trained in the natural sciences is to make models that are not only predictive but interpretable, because most biologists don’t care how well you can predict how an individual gene will respond. The biologist is inter- ested in knowing why it works that way: which parts of the genome are talking to each other; the parts of the genome where edges exist in the genetic regulatory network that they didn’t know of before.”

Wiggins is also part of a team that received a $3.7 million grant from the National Institutes of Health for creating a Nano-Medicine Center for Mechanical Biology.

“Nano-medicine is not an established field yet,” said Wiggins. “Our proposal is really an approach to build nano-scale devices to look at cellular biophysics.When you push on a cell, how does the cell know, and how does physical information become chemical information, and how does chemical information become genetic information?”

Wiggins credits information sharing through the Internet for advancing the field of computational biology — by allowing researchers around the world to share and analyze data — as well as biotechnical advances and mathematical and computational advances in machine learning.

“I’m very excited about the interdepartmental and interdisciplinary collaborations I’m working on here at Columbia, and the recognition they are now getting,” he said. “Each of these grants is very rare and special, and having two of them at once signals to the wider research community that Columbia is a leading force in biological computation.”

Related Links

Center for Computational Biology and Bioinformatics (C2B2)

Nano-Medicine Center for Mechanical Biology

Center for the Multiscale Analysis of Genomic and Cellular Networks (MAGNet)


500 W. 120th St., Mudd 200, MC 4701 New York, NY 10027 / Phone: 212-854-4457 / Fax: 212-854-8257 / Email: apam@columbia.edu

©2012 Columbia University