The Meharry dataset features diverse data from multiple, disparate sources. Thanks to a successful migration to a Meharry cloud environment, researchers will soon have access to a new powerful tool to use that data.
The EDA successfully migrated Meharry’s data to the Meharry Azure cloud environment. The transition began in Aug. 2020 and was completed in July 2023. The Azure cloud is independently hosted and managed by the EDA to support Meharry’s research enterprise. Meharry’s data was initially hosted by Clearsense.
“This is a major milestone for Meharry. By hosting the cloud itself, Meharry can also implement a data strategy that can scale to meet the ambitious goals of the College’s strategic plan”, says Ashutosh Singhal, Ph.D., chief data officer and associate vice president, clinical data stewardship.
That decision to use a third-party partner is a standard first step to adopting cloud technology.
“When companies first adopt cloud technology, they often employ a hosting provider,” says Ed Wisdom III, MBA, associate vice president, data infrastructure and analytics. “That affords them time to learn the environment and the processes involved with managing it.”
Microsoft Azure played an important role that was critical to their selection as Meharry cloud provider.
“Many cloud providers provide only the platform and tools, but do not advise you through the transition. We needed a solution that will allow us to migrate to the cloud while building our team,” says Dr. Singhal.
Microsoft provided technical expertise and guidance as the EDA developed the data architecture, completed the proof of concept that included transferring 70 percent of data, and the final production stage that implemented major security, privacy and governance features.
“The migration to the Azure cloud provides improved flexibility and scalability, adds Dr. Singhal. “We can now provide more technology and tools on the clinical operation and research side., We could not easily do that with a third-party solution.”
“We also can apply funds, that previously went to a third party, toward staff and other resources that we independently manage,” says Wisdom.
“That means we are now in the position to accept new opportunities in the machine learning/artificial intelligence area that we could not pursue previously,” adds Dr. Singhal.
Once they can access the Azure Cloud, clinical researchers will notice some considerable benefits.
“They will find a streamlined process for data storage that is both private and secure,” says Dr. Singhal. “The cloud also provides the computing power to run analytics on large data.”
The cloud environment also supports collaboration by allowing researchers to store data from various new sources.
“Researchers could combine their laboratory data with data generated by colleagues at another institution or by combining publicly available data on various cloud platforms such as human genomics data on TCGA and other sources,” says Dr. Singhal.
Currently, EDA’s clinical data includes 4M transactional clinical records and data on 220,592 unique Meharry patients. The demographics of the clinical data from the underserved and underinsured patient population is 51 percent female vs. 48 percent male; 10 percent Hispanic/Latino ethnicity; and 65 percent Black/African American.