Back in May, we (Richard Neher and I) learned that nextstrain.org was selected as a finalist for the Open Science Prize, a new initiative jointly funded by the NIH, the Wellcome Trust and HHMI. Each of the six finalists were asked to build a prototype of their project and present this prototype at the BD2K Open Data Science Symposium at the beginning of December. It was interesting seeing the other entries to the competition. As it turned out, everyone made a website. And each group was offering a layer of added value on top of publicly available data. In one example, providing a platform for sharing health and genetic information for people suffering from rare diseases and in another example, implementing a database for worldwide air quality data. A few years ago, I wrote about the possibility of a GitHub of Science. At the time, I wasn't sure exactly what this meant. I had a vague idea that someone could take a paper and fork it and add additional analyses on top of the original. Now, the future seems much more clear —
Just as software APIs allow open source software to be built layer-upon-layer, all six of the Open Science Prize finalists supply something like an API in which inputs of publicly available data are processed to yield derived outputs that encourage sharing, synthesis and understanding. I can totally imagine a scientific ecosystem in which open science projects (websites) rely on a stack of data and outputs from other groups, but produce their own data and outputs for downstream analysis. With nextstrain, we'd like to do something like this for pathogen phylogenetics and provide a basis for downstream epidemiological and evolutionary analyses. It seems like such a model could grow to live alongside the dominant (and incredibly worthwhile) scientific discourse occurring via peer-reviewed publication.
There is now public voting to determine which three entries will move forward to the final round. Although I think all six OSP entries were pretty great, we'd very much appreciate your vote. Please go to www.openscienceprize.org and vote by Jan 6.
Watercolor courtesy of Matt Cotten.