Broad Institute partners with Manifold to accelerate genomics research

The Broad Institute of MIT and Harvard is partnering with Manifold to bring a unique life science research and development platform to investigators around the world.

Healthcare data have become increasingly available in larger quantities and are stored across a multitude of cloud service providers. The Broad Institute will begin to use Manifold’s AI platform to address this problem through Manifold’s cloud agnostic architecture. Together, Manifold and Broad will adapt the existing platform to advance genomic research inside and outside of the institute.

The Broad Institute is a pioneering research lab that began when Harvard and Massachusetts Institute of Technology researchers came together in 2004 with the goal of advancing genomic research through collaboration.

“We recognized that there were certain types of research projects that were going to be really important to catalyze advances in biomedical research across the entire ecosystem, and yet it would be difficult to accomplish them in a typical academic lab,” Todd Golub, director of the Broad Institute, said in an interview.

Manifold was co-founded by Vinay Seth Mohta, now CEO, in 2016 as a consulting company for using AI in healthcare research. In 2021, Manifold began building its R&D platform to bring no-code software to research institutions.

“We started in cancer research with a focus there, because we saw that cancer consistently pushed science and technology the hardest, and also, as a result, pushed data the hardest, and where we were seeing a lot of pain,” Mohta said.

Manifold has deployed its research and development platform with the American Cancer Society (ACS) and the Indiana University Melvin and Bren Simon Comprehensive Cancer Center.

“With the Broad Institute, we're really excited about some of the novel genomics-related capabilities that will bring, based on learning about their use cases, which, by the way, are going to be the use cases everybody's going to have in about three years,” Mohta said. “Because that's what the Broad does. They're leading the world in science.”

The lessons from the Broad over its two-decade history will be incorporated into the Manifold platform. One of the key turning points from the Broad's existing software is that researchers will be able to use it without having to know how to code. The organization specifically has plans to do so with its bioinformatics pipeline.

Mohta said they plan to add new capabilities every quarter to allow more researchers to come onto the platform, along with collaborations the Broad Institute itself has with other institutions, like companies or other research institutions.

“With a more modern technical architecture, it really opens up the ability to enable AI more natively, which will again, fuel the future of discovery when we think about what we can do with the data,” Heather Jankins, co-lead of the Data Sciences Platform at the Broad Institute, said.

Perhaps the most exciting feature of the new platform, Golub said, is its “bring your own cloud” capability.

According to Manifold, "bring your own cloud" means that users of the platform can seamlessly use their existing cloud service provider to power the Manifold research platform.

“Science happens on all clouds, and so we want to make sure that we're enabling science to the maximum,” Jankins said. “From that perspective, having a platform that has natively made some cloud agnostic choices is very exciting for us to make sure that we can really leverage using the data, where the data lives.”

Research institutions use all manner of different cloud providers such as Google, Amazon Web Services and Microsoft. Without the "bring your own cloud" capability, researchers may have to consider migrating their data to another cloud to be compatible with a certain research platform or opt out of using it.

“You have this convergence of two things happening right now, which is an explosion of the amount of data that's possible to generate, and the power of AI making it possible for everybody to ask complex questions of large data sets, regardless of where they're generated,” Golub said. “That's a very different world than the way research used to work, which is that you do experiments in your own lab and you analyze those results.”

Golub continued: “This is still very early days of data generation, data aggregation, data and complex data analysis. And so we need to think about the future of this activity and what's going to be needed for the next phase, given that continued explosion of data and need, and that was a motivation for the Broad-Manifold collaboration.”

Also important to the Broad Institute was Manifold’s ability to create a trusted research environment where access to patient’s genomic data is only allowed by authorized users for a finite period of time.

One of Manifold’s current customers is the ACS, which has several ongoing cancer studies with 1.5 million participants. With Manifold, the ACS is allowing outside research institutions to use its data while ensuring the data is safely accessed.

“The point of this exercise is to accelerate the pace at which the world can make important discoveries that will help patients, and without a really scalable, durable, powerful data platform, that's not going to go as quickly as possible, and so we're excited to make that possible,” Golub said.