University researchers generate large amounts of data and therefore must be focused on its organization. The NIH Strategic Plan for Data Science, which was released last year, provides a roadmap for modernizing data science ecosystems and focuses on accessible, well-organized, secure, and efficiently operated data resources critical to modern scientific inquiry. The key components of the plan include:
- interconnectivity, and interoperability of data sets and resources,
- integration of data management tools,
- universalizing innovative algorithms and tools into enterprise-ready resources, and
- recognizing the growing costs of data management.
At the NIH Council of Councils meeting last month, an update on the NIH Data Ecosystem was given by Vivien Bonazzi — the senior advisor for data science technologies and innovation — who outlined the many challenges facing implementation of the data science plan. She shared that many programs plan to or already use cloud-based services to store and compute data, but the way data are stored and managed is unique to each program and service. This is problematic for investigators who are often unaware of how databases are constructed or do not have time to learn to use them. A clear take-home from her presentation was that developing data management plans, best practices, and use cases are a priority.
UNC investigators are also focused on data — both the science of data and practical applications and ethics of data usage. We have a collaborative community of data scientists who work to unify, maximize, and promote collective data science efforts using the NIH plan as a guideline. Changes are rapid in this ever-evolving field, and our organization must allow for innovative, flexible, and novel internal operations. To meet these needs, the OVCR, along with research deans and directors from across campus, have identified data science as an important research priority for the university. We will be sharing much more on this subject over the coming academic year.