Precision medicine is just a concept until you’ve learned enough from a slew of genomic data to make a drug. And before you even get to that point, you’ve got to store that data somewhere and organize it.
That bottleneck is a problem a number of startups are trying to solve. Today, one of those startups, a company emerging from the Personal Genome Project called Curoverse, is about to take its product to the public for the first time.
Boston-based Curoverse is beginning commercial trial runs for Arvados, an open-source computational platform that houses massive amounts of genomic data. Curoverse doesn’t “own” Arvados, but rather sets up, manages, and maintains the database for users who don’t want to install and manage it themselves. Up to this point, the service has only been available privately at places like Harvard University, where Arvados originated. Now, Curoverse will begin offering the platform to select major medical centers and research institutions and to individuals who want to tap into Arvados via the cloud.
Curoverse has set up the software at 12 centers, among them Johns Hopkins University and the Wellcome Trust Sanger Institute in the U.K. Separately, through a public beta program, individuals can log on to the company’s website, create an account, and try out the software. That trial run will last six months, and it gives folks access to 1 terabyte of free storage, 100 hours of free computing time, and all the public data shared on Arvados, according to Curoverse co-founder and VP of customer and business development Jonathan Sheffi.
Curoverse is using this test run to work out any remaining kinks, and see how potential customers take to the service. The startup plans to roll the database out commercially over the summer, at which point it’ll start charging users a subscription fee for usage (CEO Adam Berrey won’t say how much the company will charge).
“The software has been used a lot, so we feel very confident in its stability,” says Berrey (pictured above, left, with Curoverse chief scientist Alexander Zaranek). “It’s really ready for a larger trial, and larger adoption.”
The move is the latest step for Curoverse, a startup that emerged from George Church’s Personal Genome Project at Harvard. The PGP was a plan led by Church to sequence more than 100,000 genomes in the U.S. and link them to individuals’ health information. (The same kind of aggregation, but of 1 million people’s genomic and other health information, is a goal of the Obama administration’s Precision Medicine Initiative.) Church needed a massive database to house all that information, and that’s what led to the creation of Arvados. It’s a database capable of storing giant amounts of genomic information, it’s shareable, it can run on both public and private cloud services, and it’s an open source platform, so anyone can use or modify the source code.
Curoverse points to this last fact as key to differentiating Arvados from competing genomic database tools like DNAnexus, made by the Redwood City, CA-based company of the same name. “A community forms around open source projects where everyone’s adding to it, and everyone’s improving it collectively so it becomes bigger and stronger than any one institution or any one vendor,” Berrey says.
Curoverse raised $1.7 million in seed funding from backers such as Hatteras Venture Partners and Point Judith Ventures in December 2013. Berrey says those backers have poured more into Curoverse since then—more than the $500,000 outlined in a regulatory filing in late March—and that the company aims to move on to a full-fledged Series A round once it officially launches Arvados later this year.
Berrey says since Curoverse began putting its seed round together in 2013, interest in precision medicine—more targeted therapies based on genomic and other data—has only increased. One major question facing Curoverse when it was conceived, for instance, was whether a need would emerge for the type of software it was developing. Berrey says programs like the Global Alliance for Genomics & Health, a two-year-old international coalition trying to establish a common framework for genomic data sharing, have helped answer that question affirmatively for Curoverse.
“That is 100 percent aligned with the vision we laid out,” Berrey says of the initiative. “The timing now is perfect.”
The buzz around precision medicine generated by President Obama’s proposed initiative to build a massive database of genomic data has only helped. As Berry says, it’s “aligned everyone’s thinking around the way medicine is evolving.” Curoverse will try to capitalize by helping researchers store, organize, and share data with one another through Arvados. It’s targeting medical centers, pathology and independent genetic testing labs that use sequencing tools for diagnostic tests, and drug developers, among others, as potential customers.
“The pilot program gets the software into the field and gets people using it,” Berrey says. “[Then] we’ll start to scale deployment.”