ENGINEERING

Sophisticated Algorithms Analyze Data to Develop Predictive Models of Disease in order to Better Diagnose and Treat Patients

Ushering a new era of “precision medicine,” Mount Sinai is one of the first academic medical centers in the United States to build and operate a supercomputer. Named after the Roman goddess of wisdom and medicine, “Minerva,” uses sophisticated supercomputer algorithms to analyze data in order to develop predictive models of disease that can better help diagnose and treat patients.

After a full year of operation, the $3 million computer is a large step forward in building a significant computational and data-intensive infrastructure geared specifically for genomics. Built on-site by Mount Sinai’s Department of Supercomputing, Minerva analyzes the growing “digital universe”, including genomic and phenotypic data, as well information from electronic medical records (EMR). In fact, it has already analyzed hundreds of human genome sequences with a projection of hundreds of thousands every year. With thousands of processing cores working in unison, and tens of terabytes of memory, the supercomputer can perform complex and sophisticated tasks quickly and with more precision than ever before.

“With Minerva, Mount Sinai has the ability to quickly analyze genomic patterns to provide a greater understanding of the causes of disease and how to personalize treatments according to an individual’s genetic composition,” said Dennis S. Charney, MD, Dean of the Icahn School of Medicine at Mount Sinai. “The supercomputer is able to accomplish real-time visualization of advanced molecular models, promoting drug development and allowing us to test the effects of molecular variations on different receptors in the body.”

The supercomputer also stores information from Mount Sinai’s biobank, or BioME, a collection of over 24,000 individuals’ DNA and plasma samples that are stored in a way that protects patients’ privacy while allowing research to be performed. Funded by the Charles R. Bronfman Institute for Personalized Medicine, the biobank accesses a broad range of genetic and environmental information on patients who have agreed to lifelong participation.

“The supercomputer is helping us better understand and foresee the course of disease for each patient -- and to identify the outcome to a particular therapeutic intervention in advance,” said Patricia Kovatch, Associate Dean for Scientific Computing at Mount Sinai Medical Center and the engineer who constructed Minerva. “Thus, using genomic data, information from our biobank as well as complex simulations of molecules, we are able to enhance personalized medicine to a degree that has never been done before.”

Eric Schadt, PhD, Mount Sinai’s Director of the Institute for Genomics and Multiscale Biology, cited the need for Minerva in critical areas of research that is already underway. “In order to analyze and integrate all the different data dimensions over the population, and build predictive models of disease, we need the supercomputer. With the infrastructure we’re creating, and the people we’ve recruited, combined with the resources already available at Mount Sinai, we are coming together to form a new epicenter of research on personalized medicine and the new biology.”

“Few research centers have the type of computing infrastructure to allow advanced modeling that Mount Sinai can now do on-site. Along with other advances in genetics and some recent outstanding additions to our faculty, Minerva further cements Mount Sinai’s reputation at the forefront of the ‘precision medicine’ movement,” said Dr. Charney.

Joseph Buxbaum, PhD, Director of the Seaver Autism Center for Research and Treatment, says Minerva’s supercomputing capacity plays a key role in analyzing data gathered as part of the Autism Sequencing Consortium, a multinational collaboration where all the 22,000 genes in humans will be sequenced in thousands of individuals with autism spectrum disorder. “The consortium plans to have such data for as many as 30,000 people – including controls – over the next three years, and we anticipate that this will result in the discovery of several hundred autism genes. Without Minerva’s computational power, a project of this scope would simply not be feasible.”

To data, tens of millions of core processing hours of work has been done by Minerva, added Kovatch. “The [super]computer has helped scientists publish over 25 research papers. Minerva helps scientists analyze their data quicker than ever before, as well as complete more complex tasks simultaneously. The end result is that more science, even basic science, is done quicker and more efficiently.”