Heart organization building big data platform to aid precision medicine
The American Heart Association is striking out on an ambitious plan to use big data analytics, data mining and the cloud to enable the use of precision medicine to treat cardiovascular disease.
Leading the way for the AHA is the Institute for Precision Cardiovascular Medicine, which is coordinating the national organization’s initiatives. The institute was launched last October and given a charter to develop a long-term plan to use precision medicine in the battle against heart disease.
In its first year, the Institute has pursued an approach where it’s serving as a convener, drawing together organizations interested in precision cardiovascular and providing grants for research.
However, the AHA has bigger ambitions to facilitate research and is looking to develop a “data discovery platform” that will enable information sharing, research and collaborate in the field.
Toward that end, the Institute is researching open-source components and looking to the cloud to support data sharing—in July, it formed a strategic partnership with Amazon Web Services (AWS). However, it’s not an exclusive relationship, and the Institute would be open to using other cloud services, such as those offered by Microsoft or Google, to promote data sharing on the scale it’s envisioning.
While the deal with Amazon is a first step, the Institute’s data platform is still very much in the design phase, says Taha Kass-Hout, an AHA strategic advisor for data and health informatics.
The Institutes hopes to create both a social and computational infrastructure that will ease data sharing and collaboration among researchers. “We’re trying to focus on the platform as an open stack that will enable search and discovery, and harmonization of the data,” says Kass-Hout, who was the first chief health informatics officer for the Food and Drug Administration and the creator of precisionFDA, an online, cloud-based, portal that enables scientists from industry, academia, government and other partners to come together to foster innovation and develop the science behind a method of “reading” DNA known as next-generation sequencing.
Precision medicine is a fast-growing approach to disease prevention and treatment that seeks more precise care by combining scientific research with detailed information about a patient’s genes, environment and lifestyle. Key to support such efforts is collecting vast stores of health data that researchers can analyze to identify patterns and specific ways to diagnose and treat individuals.
The AHA has high hopes for precision medicine and believes building an open research platform through the Institute will increase opportunities to use personalized medicine in treating cardiovascular disease and stroke. The organization allocated $30 million over five years to launch the Institute, and a fund-raising initiative is underway to generate another $100 million to $200 million for future work.
“Tailored prevention and treatment will help patients and lessen the global burden of cardiovascular disease,” says AHA CEO Nancy Brown.
In addition to providing leadership and funding support for the development of precision cardiovascular medicine, the Institute is expected to catalyze, coordinate and integrate multiple activities across big data platforms that it hopes will lead to the generation of the new data discovery portal. And it will bring stakeholders together and serve as an “honest broker” to balance the interests of competing parties, says Brown.
In the near term, the Institute will include data from paired genotype-phenotype datasets—such as longitudinal studies and health systems biobanks—to help provide access to a large, highly diverse set of the most valuable cardiovascular data for researchers. In the future, it will include a much wider set of cardiovascular-related data, including information from registries, patients, various molecular and genomic studies, electronic health records, clinical trials and regulatory databases.
Kass-Hout expects that the eventual platform will be used to “build an environment and ecosystem to advance science through collaboration.” Data sharing and collaborative work on research will provide “the environment, space and focus on cardiovascular information that will help advance the science.”
He expects that the institute’s data platform will contain a variety of data manipulation and research tools that will support analytics and data harmonization. “Our whole intention is to provide the platform as an open-source environment,” Kass-Hout says.
While data on the platform will come from a variety of sources, the AHA intends to ensure that it won’t include patient identifiable information, Kass-Hout says.
The eventual infrastructure to support the Institute’s vision for collaborative research will require massive capacity to enable collaboration in research. Entire genomic sequences for humans can require 200 gigabytes of storage space for each person, so any study involving hundreds of humans can require significant amounts of storage to hold all the data.
Kass-Hout acknowledges that the emerging platform will evolve over time to both hold the massive amounts of data and enable researchers to work on finding insights from it. “It will require the ability to do data crunching, and Institute wants to develop a platform that has that capacity for those who might need the platform as a way of conducting the research,” he says.
Because of the large amounts of data involved and the need to share it with far-flung researchers, the AHA is stressing the development of cloud-based approaches to facilitate investigation.
“The cloud is not just a technology to us, but also a social and ecosystem enabler,” Kass-Hout says. “Many institutions already have data in the cloud, and it’s a tool to help us scale our efforts. We’re thinking about both storage and computation, which would occur in an environment with multiple cloud partners that can support high-performance computing.”
The initial partnership with Amazon AWS is a way for the AHA to grab a toehold in cloud-based research while determining how it will conduct such efforts in the future.
Brown says the collaboration takes advantage of both entities’ strong suits—the technological capabilities of AWS and the scientific and convener roles of the AHA.
By enabling the dissemination and sharing of vast quantities of genomic and other data, the AHA will be able to achieve its Institute’s goals of providing access to information to support the work of researchers around the world who are studying precision medicine approaches to predicting and treating cardiovascular disease, and that can lead to useful discoveries. “Those who know technology, data and bioinformatics and can provide important new knowledge in cardiovascular research,” she says.
Meanwhile, the capacity of the Amazon cloud service will assist research efforts on heart disease, says Teresa Carlson, vice president of the worldwide public sector for AWS. “We’re positioned to provide scalable, cost-efficient solutions for the scientific community, while delivering the industry-shaping technology and high-performance computing necessary to facilitate demanding research projects.”
The Institute’s latest grant portfolio includes more than $2 million in cash awards, provided by the organization, and $2.6 million in technology credits from AWS; those credits can be used by grant recipients for free access to AWS services, which can include computational storage and analysis on its cloud.
Through these grants and fellowships, the AHA hopes to enable the scientific, mathematics and technology community to discover solutions to overcome the current obstacles in accessing and utilizing data. The recent grants announced in conjunction with the Amazon AWS partnership include:
- Data mining grants are aimed at uncovering patterns and knowledge within existing data sets that will inform standards and protocols for organizing and categorizing current and future cardiovascular disease-based data. These grants will also focus on data harmonization efforts necessary to enable data mining.
- Methods validation grants will focus on validating pre-existing methods including algorithms and analytic tools used to maximize the use of data in predicting outcomes.
- Innovative development grants will focus on developing tools that enrich the ability to identify novel approaches and/or tools to analyze data.
In addition, the AHA is planning fellowship awards that are intended to cross-train scientists interested in learning computational biology or more information about specific cohorts. “The data fellowships aim to establish a future generation of researchers with enhanced skills and capabilities,” Brown says.
“The AHA’s research has always been evidence-based, and with the emerging work of the Institute, it’s going after providing that evidence in for precision medicine,” Kass-Hout added. “Researchers will be able to have access to high-resolution data, like genomic studies. We’re looking forward to the next generation of data that will enable early diagnosis and prediction; this provides a wonderful opportunity to look at systems biology. Through coordinated research, we’ll be able to develop different ways of looking at this knowledge, and it’s a chance to redefine our understanding of cardiovascular disease.”