IBM, Revolution Analytics Speed Up MS Research at SUNY Buffalo

The State University of New York at Buffalo is using an IBM Netezza appliance and Revolution Analytics software to create algorithms to accelerate multiple sclerosis research.

The State University of New York (SUNY) at Buffalo is developing algorithms using an IBM Netezza 1000 data warehouse appliance and Revolution Analytics software to further multiple sclerosis (MS) research.

The algorithms will allow the university to study genomic data sets for more than 2,000 genetic and environmental factors that may lead to MS symptoms. IBM revealed details about the genomic research project on April 26.

Personalized medicine using big data is a major trend in health care, and the IBM hardware and Revolution R Enterprise software allow researchers to develop individual treatment to slow MS symptoms such as brain injury, physical disability and cognitive impairments, IBM reported.

Researchers will examine how factors such as gender, geography, ethnicity, diet, exercise and sun exposure may lead to MS. Living and working conditions may also lead to the disease, IBM reported.

"My group has been interested in developing new algorithms and new methods for gene environment interactions," Dr. Murali Ramanathan, lead researcher at SUNY Buffalo, told eWEEK. "We have used Netezza appliances along with Revolution R capabilities to speed up and enhance the performance of our gene environment interaction and analysis algorithms."

Before using the Netezza appliance, analysis of genetic material would take several days, and now it takes only a few minutes, said Ramanathan.

"We were able to do thousands of permutations within the Netezza box, something we were not able to do in our earlier implementation," said Ramanathan. The increased speed and performance allows researchers to explore additional questions on the role of genetics and environment in leading to the onset of MS.

In addition, the Revolution R software allows researchers to streamline their workflow to handle a huge amount of the genetic data, according to Ramanathan.

"In the past, the SUNY team would have had to rewrite the entire algorithm, which would have required a great deal of time from a grad student or Ph.D. candidate," David Smith, vice president of marketing and community for Revolution Analytics, told eWEEK in an email.

Using Revolution R Enterprise for IBM Netezza, SUNY Buffalo researchers were able to reduce computation time from 27.2 hours without the Netezza device to 11.7 minutes with it, said Smith, who wrote about the project in a blog post.

"With Revolution R Enterprise for Netezza, the R language is embedded in each of the processing cores of the IBM Netezza 1000 appliance," said Smith. "This not only enables massively parallel computations using the R language on the high-performance IBM hardware, it also means that data does not have to be moved to another environment for processing, [therefore] further increasing performance."

About 400,000 people in the United States have been diagnosed with MS, according to the National Multiple Sclerosis Society.

IBM BladeCenter servers power the Netezza storage appliances that SUNY Buffalo is using.

Although SUNY Buffalo has been working with Netezza appliances for two years, the addition of Revolution R software brings added analytics capabilities to study more interactions between genetic material and environmental factors, which are called phenotypes. "Now they can do over 1,000 variables getting more of these other kinds of data," Shawn Dolley, vice president of big data, health care and life sciences at IBM, told eWEEK.

"[Groups] of MS patients with phenotypes in common are what lead to more narrowed interventions," said Dolley.

The goal of the research is to not only develop a cure for the disease but to also help MS patients have quality of life, Dolley noted.

"The technology and the algorithms helped us address risk and progression," Ramanathan added.

"The collaborative environment and the new algorithms and architectures have been very productive, and our hope is that we'll be able to leverage these new algorithms and data sets for developing a cure and prevention for MS," said Ramanathan."That's our hope and mission."