NEW ORLEANS (GenomeWeb) – At the American Society for Microbiology meeting here yesterday, researchers sought community feedback on what shape a National Microbiome Data Collaborative might take.
There's been a surge in the number of microbiome datasets produced in recent years. At ASM, the University of Maryland's Claire Fraser noted that the number of microbiome datasets increased from almost nothing a few years ago to millions of shotgun sequences available today. She added that the numbers could increase by some 100- to 1,000-fold in the next decade.
"Our ability to generate sequences … has far outpaced the infrastructure available," Fraser, one of the co-chairs of the town hall session to discuss the idea, said.
"We have an awful lot of room for expansion," she added.
Microbiome data spans a range of fields, including energy, health, food production, and environmental stewardship applications, the University of Hawai'i at Manoa's Ed DeLong noted during the session But it also requires some common approaches. "We all know the microbiome is important in many contexts," said DeLong, one of the session co-chairs.
At a small workshop at the Department of Energy's Joint Genome Institute earlier this year, a group of researchers, including DeLong and Fraser, discussed how to advance the microbiome field and developed the idea of a National Microbiome Collaborative, which they presented at ASM.
The goal is to form an integrative collaborative so that microbiome data that might otherwise be lost could be captured, while also providing access to quality datasets and reference databases. At the same time, it could set standards for the field at large and provide analysis and other tools.
For such an effort to work, Fraser and DeLong said the broader community has to be involved. "This is not going to happen unless there is buy-in from interested stakeholders," Fraser said.
But there wasn't a clear consensus from the packed room on how to proceed.
One audience member suggested starting from square one and developing an entirely new database, while others suggested building on previous efforts by the Human Microbiome Project or the National Center for Biotechnology Information, as well as learning from other disciplines like physics that devised way to cooperate and share data.
At the same time, a number of audience members argued that datasets needed to meet certain minimum standards and provide a certain base level of metadata about how and where samples were collected to be useful. A collaborative, they added, would offer an opportunity to develop best practices for the field.
Another audience member, though, proposed not keeping all microbiome datasets, but instead developing a collection of curated reference sets.
Fraser and DeLong encouraged the audience to continue to contribute their ideas via email, and said that they would be presenting the idea to an even larger group in the fall.