
A Comprehensive Platform for Agri-Food Data Analysis
Set Apart from Other Analytic Platforms
Enabling the discovery of new insights, new partnerships, and new interconnections among datasets.
GEMS makes it possible (easy, even) for data scientists to create research-ready agri-food data, turn it into actionable information, and share it with partners (public and private) in a trustworthy environment. How GEMS Enables Data Discovery
While other platforms provide data in a variety of incompatible formats, the GEMS Platform provides tools to standardize metadata via internationally-curated agri-food ontologies and controlled vocabularies, so that data can become interoperable with other datasets. This metadata can be published separately in the platform so that others can discover your data without access to the data itself, until you provide them permission.


Trusted Security And Smart Sharing
Full IP protection with progressive protocols, designed for sensitive data, coupled with the intelligent option to share selectively.
Although GEMS was hatched in a public land-grant university, it was conceived from the beginning to fully protect IP, whether that be pre-publication data of university scientists, identifiable attributes of family farms, or corporate data assets. How GEMS Protects Your Data
The GEMS Platform is hosted at the Minnesota Supercomputer Institute, which also handles sensitive personal health data, and has state-of-the-art technical security protocols in place. Further the data housed within GEMS has legal privacy protection due to a Minnesota state government statute that came into effect in August 2018. Users can also opt to make their data and workflows fully public in GEMSOpen or other platforms, as they choose using Smart Sharing capabilities.
Collaborative Analytic Tools
For data science teams ready to take action, GEMS offers a highly organized, synchronized environment to programming.
The GEMS platform includes an analytic environment that pairs versioned data with versioned tools to create reproducible workflows. Its team-centric configuration allows for the collective processing of agri-food data into actionable insights. How GEMS Enhances Data Science
GEMS’s Jupyter environment goes a step further than other seemingly similar environments in that all analyses operate in a filesystem synchronized with a sandbox workspace, an RStudio interface and a desktop environment. Full support in Jupyter is provided for programming in R and Python, the lingua franca of contemporary data scientists.
