Strata Rx 2013: Data liquidity drives disease research

BOSTON—“We don’t have big data, but gazillions of pods of little data,” said Marcia Kean, Feinstein Kean Healthcare, at Strata Rx 2013 on Sept. 26. However, data liquidity—or the rapid, seamless and secure exchange of useful, standards-based information among authorized individuals and institutional senders—is poised to significantly advance biomedical research, she said.

Data liquidity can enable the integration of research studies from disparate organizations and companies, speakers said.

At the Accelerated Cure Project (ACP) for Multiple Sclerosis (MS) in Waltham, Mass., a data repository of bio samples, including DNA, RNA, plasma and serum from 3,200 patients with MS, is available for research. This repository contains medical history, demographics and a 52-page case report for each of these patients.

“This has supported 77 research studies worldwide and it will support hundreds more,” said Robert McBurney, CEO, ACP, noting that they cover MS treatments/therapeutic targets, risk factors, genetics, diagnostics, immunology and disease properties. To use the repository, organizations are required to return data on their findings.

With more than 70 datasets across multiple dimensions, hundreds of samples were used in multiple studies. This powerful overlap of datasets offered an opportunity to revolutionize MS research, said Dave King, Exaptive.

For example, so far six studies use data from the same 164 people with MS.

“The importance of this was to create a tool for researchers to see how their research fits into a broader landscape,” he said. Meanwhile, the tool sparks collaboration between the for-profits, non-profits and academic sources who have used the data in their research.

“It’s real collaboration with what started off as serendipitous collaboration,” King said.

Kenneth Buetow, director of computational sciences and informatics within Arizona State University’s (ASU) Complex Adaptive Systems Initiative (CASI), focuses on using computational technologies to solve biomedical challenges, in particular the role of genetics in cancer.  

At ASU, he utilizes heterogeneous cancer data, including molecular data, image data, biologic process data, gene-disease relationship data and gene-drug relationship data, to identify trends in survival outcomes with different interventions. This should inform physicians at the point of care, he said.

By looking at the impact of genomic profiles, “at least it lets you leverage what is known in a data-driven manner.”

He stressed the importance of data connectivity and the need for specialized entities to share their data and studies. For example, researchers studying cardiovascular disease, type II diabetes, liver disease and cancer too often work in silos, he said. Even accessing published genetic study data is a challenge, and these walls will need to be torn down to truly drive research.