Editorial: Why keeping big data free matters to me—and us all

I began my master’s program in biological sciences at the ripe old age of 21 with a small amount of undergraduate research under my belt and a dream of studying bioinformatics. The laboratory I joined was small, just one PhD student and two master’s students, including myself. And there just wasn’t enough funding for everyone to generate their own new data.

So my adviser introduced me to the US National Institutes of Health National Center for Biotechnology Information’s Sequence Read Archive (SRA) and said everything I needed to complete my thesis was in there.

The SRA, as I would learn, is a massive repository containing over 12 petabytes of data generated from sequencing experiments. And it is completely free to access.

The federal government didn’t pay me a dime to do research during my master’s. But you could say that every dollar it spent to support DNA-sequencing-based research elsewhere was also spent to support me. And I think that’s a model that should be used to support more young scientists all over the world.

Technological advancements over the past several decades have enabled scientists to generate more and more data. Now the problem is that we have more data than we know what to do with.

So roughly 10 years ago, scientists who understood we needed better ways to handle this explosion of data created the FAIR principles. These principles of findability, accessibility, interoperability, and reusability helped establish guidelines for data curation. At the 2016 G20 Summit on Financial Markets and the World Economy, the global community endorsed the FAIR principles as a guiding beacon for sharing scientific data. And both public and private data repositories began adapting their data submission requirements to these principles.

Research metadata is now somewhat standardized, and researchers can find data relevant to their research more easily than before. That’s a huge boon for the scientific community.

Beyond the SRA, there are countless other massive data repositories, including ones that hold bioactive chemicals, crystal structure data, or protein sequences and structures. For nearly every scientific discipline, there exists a database out there somewhere to explore. And private industry assuredly has their own datasets in need of exploration too.

Without a doubt, there exists within those data repositories discoveries to be made that can rival those that require millions of dollars in new investments.

But these data repositories shouldn’t be considered permanent infrastructures, especially given the massive cuts to science funding happening in the US right now. Redundancy may not be one of the FAIR principles, but researchers should consider making their data available in multiple places and publishing the locations of their data on their personal lab websites. Because if a disruption to these systems causes a link to data in an old paper to break, there needs to be some alternative way of helping people find it.

These data-use suggestions are not meant to diminish the impact of the recent firings and grant cancellations in the US. Research using previously generated data in no way replaces the new research now put on hold or the expertise lost from these scientific agencies.

But we still need to think about how to train the next generation of scientists and continue on with research. Using freely available data could be one way to achieve that.

This editorial is the result of collective deliberation in C&EN. For this week’s editorial, lead contributor is Max Barnhart

Views expressed on this page are not necessarily those of ACS.