Skip to main content
Premium Trial:

Request an Annual Quote

FDA Taps DNAnexus to Build Communal Infrastructure for NGS Test Development

Premium

NEW YORK (GenomeWeb) – With support from an $849,000 research and development contract from the US Food and Drug Administration's Office of Health Informatics, DNAnexus will develop and help pilot a platform-as-a-service for sharing genomic data and tools and generating reference standards to support next-generation sequencing-based genetic tests.

The open-source platform, called PrecisionFDA, seeks to draw test developers, standards-making bodies, pharmaceutical and biotechnology companies, healthcare providers, and academic medical centers into a collaborative forum where they can assess the quality of testing methods and help derive standards around secondary analysis of genomic data, David Shaywitz, DNAnexus' chief medical officer, told GenomeWeb.

Echoing similar sentiments in his comments, Taha Kass-Hout, FDA's chief health informatics officer and director of the Office of Health Informatics, described the system as an effort to bring relevant stakeholders into a single environment and enlist their help in both deriving and honing meaningful standards. Ultimately, the goal is to help the FDA figure out which standards should be in place to ensure test quality and accuracy and how those standards should implemented in keeping with existing policies, he told GenomeWeb.

This is especially timely in light of the recently launched Precision Medicine Initiative, which aims to use NGS technologies to personalize treatment and care on a large scale. As part of efforts to enable that vision, the FDA is seeking to evaluate whether test developers should be required to show that their tests meet certain quality standards, Kass-Hout and colleague David Litwack, who is a policy advisor with the Office of In Vitro Diagnostics and Radiological Health in the FDA's Center for Devices and Radiological Health, wrote in a blogpost on the FDA's website.

The FDA is also trying to assess whether it would be more efficient for developers to reference shared repositories of validated and curated mutations to assess the clinical value of their tests rather than try to generate mutation-disease associations themselves, they wrote.

PrecisionFDA is an opportunity to explore the efficacy of those ideas in practice and to create the requisite data and materials. Cloud infrastructure is well suited for this sort of tire kicking because it can handle multiple users and scales to meet rising demands for compute power and resources, Kass-Hout noted. It's also easier to track community activity in the cloud and keep abreast of new developments, he added.

DNAnexus, for its part, has been a long-term player in the genomics space and can boast of an established community that is familiar with its platform and confident of the firm's ability to ensure the security and privacy of uploaded data, which is why the FDA chose to work with the company, he said. The agency did consider alternatives — including open-sources system, such as Galaxy, as well as some unnamed proprietary systems — and it is keeping its options open, Kass-Hout said, but DNAnexus' system best fit the bill in terms of the "maturity [of the system] ... and also given the timeline [within] which we want to showcase our approach."

The partners plan to release the first iteration of PrecisionFDA for testing as part of a closed beta in December this year. During this beta-testing period, the FDA will work with a select group of academic laboratories and centers and standards-generating bodies to evaluate the efficacy of the system and community model. Down the road, "hopefully, that will expand to include patient consortia, providers, and other agencies," Kass-Hout said. The focus for the beta might be on hereditary diseases testing, but the FDA has not made the final decision on that front, Litwack said.

The data and all the secondary analyses — sequence mapping, alignment, and variant calling — will be hosted and run on DNAnexus' platform and will be accessible through a web-portal that the company is developing specifically for PrecisionFDA, Omar Serang, DNAnexus' chief cloud officer, told GenomeWeb. The portal will include tools for assigning administrative access to different community members as well as tools for managing projects and reference datasets, samples, and pipelines, he said.

The system will also include applications, such as a so-called FASTQ injector, that will allow users to make modifications to FASTQ files and test how well their pipelines are able to pick up these additions, Serang said. It will also offer a call comparator tool that will let users compare submitted variant call sets to reference datasets generated by standards organizations such as the National Institute of Standards and Technology and its Genome in a Bottle Consortium

Community members will also be able to contribute to the development of new reference datasets. One of the planned activities for the community, Shaywitz told GenomeWeb, is to select a set of samples and have multiple members do deep sequencing of these samples, analyze the results, and see what variants their methods are able to identify. The community would evaluate and compare the results and try to arrive at a consensus dataset. DNAnexus is developing a mechanism by which community members will be able to vote on which specimens should be selected for this purpose, Serang told GenomeWeb.

Within the platform, community members will be able to share analysis results with participating collaborators and the FDA as well as share pipelines and software code with the community if they choose to. Members will have access to private spaces within the system where they can store their datasets and tools, and they'll be able to control who is able to access these resources, Kass-Hout said. There also will be separate spaces for collaboration as well as a public space that will host a project wiki and a question-and-answer forum, among other resources. Part of the beta will involve determining how much private space and compute resources users will have access to, Kass-Hout said.

There are no costs associated with joining the community, according to the FDA.

The Scan

Push Toward Approval

The Wall Street Journal reports the US Food and Drug Administration is under pressure to grant full approval to SARS-CoV-2 vaccines.

Deer Exposure

About 40 percent of deer in a handful of US states carry antibodies to SARS-CoV-2, according to Nature News.

Millions But Not Enough

NPR reports the US is set to send 110 million SARS-CoV-2 vaccine doses abroad, but that billions are needed.

PNAS Papers on CRISPR-Edited Cancer Models, Multiple Sclerosis Neuroinflammation, Parasitic Wasps

In PNAS this week: gene-editing approach for developing cancer models, role of extracellular proteins in multiple sclerosis, and more.