As an open repository, SPARC provides resources to share and utilize public datasets. The SPARC ecosystem consists of several integrated platforms and tools developed by an interdisciplinary, international team of data experts. The SPARC ecosystem is developed with scale and sustainability in mind. This document outlines some of the efforts and guarantees that SPARC has put in place to assure investigators that SPARC is the right repository for their scientific data for the long term.

Guaranteed Data Availability

SPARC guarantees that data submitted to the SPARC repository will continue to be available for a minimum of 10 years post publication. SPARC assigns each dataset a Digital Object Identifier (DOI) that will resolve to the location of the dataset. As outlined below, SPARC has a diversified funding portfolio to ensure continuous operation after funding from the SPARC program ends. To provide extra guarantees that submitted data will be available for the long term.

SPARC works closely with the NIH Strides initiative (STRIDES Initiative | Data Science at NIH) for long term data storage for submitted datasets. In addition, the University of Pennsylvania guarantees availability of the data for 10 years post publication in case other efforts are no longer funded. In this scenario, data will continue to be made available through the University of Pennsylvania, and DOIs will be updated to continue to point to the datasets.

Sustainable platform technologies

SPARC comprises a number of platform technologies such as SciCrunch, Pennsieve, o2S2PARC, and the MAP-Core tools maintained by the four SPARC DRC Cores. Each platform is developed to support multiple projects and is therefore funded through multiple funding sources and mechanisms. SPARC leverages this to ensure continuous operation of the platforms.

Open source projects

Key components of the infrastructure are developed as open source projects. Their code continues to be publicly available, even if funding should run out. Running and maintenance of the platform could therefore be picked up by the community.

This concerns the following components:

• SPARC Portal: https://github.com/nih-sparc
• o2S2PARC:
  • https://github.com/ITISFoundation/osparc-simcore
  • https://api.osparc.io/dev/doc
• Pennsieve:
  • https://github.com/Pennsieve
  • https://docs.pennsieve.io/reference
• MAP tools
  • Flatmaps / knowledge management are primarily using the repositories under the https://github.com/AnatomicMaps organization.
  • Scaffold mapping tools, described in this tutorial for getting started: https://github.com/ABI-Tutorials/ABI-MappingTools
  • Portal viewers / components, while used in various places on the portal, are all covered by the Map Integrated Vuer: https://github.com/ABI-Software/mapintegratedvuer
  • Simulation tools (used in o2S2PARC and via the portal) are located in the https://github.com/opencor organization.
• K Core:
  • SCKAN SPARC Knowledge Graph: https://zenodo.org/records/7476115
  • SciGraph: https://github.com/SciGraph/SciGraph
  • SPARC Ontologies: https://github.com/SciCrunch/NIF-Ontology
  • SDS and SCKAN Curation Tools: https://github.com/SciCrunch/sparc-curation
  • SDS viewer: https://github.com/MetaCell/sds-viewer
• Other Contributors
  • FAIR Data Innovations Hub, a division of the California Medical Innovations Institute (CalMI2)
    • SODA SDS Tool: https://github.com/fairdataihub/SODA-for-SPARC