Understanding Transplant Rejection and Infection

Dr. Chris Larsen is an Emory transplant surgeon and immunologist who is partnering with the AWS at Emory team to achieve proactive patient care.

A little background about transplant work: The goal of Dr. Larsen’s team is having one transplant last a lifetime. While most patients do well with their transplants, there are those whose bodies get infections or reject the transplant completely. To aid in transplant acceptance, Larsen and his collaborators developed an immunology-suppressive drug (Belatacept) which increases transplant survival rates.

By taking advantage of metagenomics, Larsen’s researchers perform deep analyses on the way viral infections attack the immune system. In doing this research, the datasets become huge. This is where AWS at Emory has been a crucial partner.

The AWS at Emory team helped Dr. Larsen build a cloud computing platform that enables the Emory Division of Transplantation to analyze both clinical data science and bioinformatics systems immunology.

“It has been wonderful,” says Dr. Larsen. “They have helped us understand the resources and easily launch complex and computationally intensive jobs. The partnership with both the AWS at Emory team and the consultants from Amazon helped us build the analytic pipelines and create solutions that are easier to demonstrate.”

As a trained biologist, Chris had always relied on small groups for analysis. Now he and his team collect gigabytes of data as they follow several thousand transplant patients for life to help understand the outcomes of transplant rejection and infection.

Through the Clinical Data Warehouse, millions of labs, medications and hospitalizations provide the data that gives his researchers a real picture of what is going on with patients. This level of computational work creates computationally intensive results that would be impossible on local machines.

“The standard procedures for care have been to treat infections as they arise. We’ve never had predictors or a monitoring scheme to detect the likelihood of problems,” says Larsen. “Now we can take a more preemptive approach.”

After years of building a clinical data mart to understand transplant outcomes, in collaboration with the OIT Data Solutions team, and then marrying that data with an AWS-based computational platform, Dr. Larsen’s team is predicting and identifying the subsets of people who are at risk and creating new immuno-suppressive protocols to help them.

“This is a change in clinical practice that was driven by our access to this data and our improved ability to analyze it,” says Dr. Larsen. “That’s exciting to me. I would encourage my colleagues to undertake similar collaborations with AWS at Emory. The scientific advantages of partnering with our OIT teams are very powerful.”