Facilitating the Development of Modular Data Models in Materials Science

Summary

The Materials Data Curation System has the potential to accelerate interoperable data exchange within materials science and engineering. This platform is designed to enable the typical PI to easily disseminate diverse datasets with sufficient domain-specific metadata through various access methods. In order to use the Materials Data Curation System, the user must compose a “Template”, which is a comprehensive description of the structure of metadata, data, and files under consideration, within their unique project. Although the Materials Data Curation System Composer allows the user to create a Template from scratch, this approach results in redundant efforts and limited interoperability. An alternative approach would be to develop and disseminate broadly applicable data models, which could be assembled, at the user’s discretion, into Templates. This approach both decreases the burden on the user and increases the interoperability and discoverability, while still providing flexibility for users to define a unique Template. The focus of this project is to coordinate effort and convene experts to address relevant challenges. We are currently focused on three areas:

• Development of foundational data models for concepts used throughout the field
• Development of modular data models for specific modeling, synthesis and measurement techniques
• Facilitating integration of the Materials Data Curation System within new and established research projects

This effort should reduce the barrier of adoption among new users and significantly expand the breadth of materials data currently discoverable and accessible.

Description

Development of Data Models

This project is convening experts with the goal of developing foundational data types and domain-specific data models to address common modeling, synthesis and measurement techniques. The existence of such data models will facilitate the adoption of the Materials Data Curation System. Current focus areas for foundational data types include representations for physical quantities, uncertainties, Miller indices, and tensor quantities. Current focus areas for domain data models include atomistic simulation (classical and quantum mechanical), diffractometry, and electron microscopy. The community is encouraged to modify and expand upon these foundational data types and domain data models. To make such community-supported data models discoverable and accessible within the broader Materials Innovation Infrastructure, a repository is being developed.

 

Integration with Existing Research Projects

To demonstrate the use of the data models described above, this project is supporting integration in the following research projects:

• NIST Interatomic Potentials Repository
• High-Throughput (Combinatorial) Materials Science
• Cloud of Reproducible Records
• Generic Materials Design Toolkit
• DFT Benchmarking
• Grain Boundary Properties

Would you like to get involved?

Would you like to know more or participate in data model development?

Would you like to attend a new users workshop?

Would you like to help coordinate a domain-specific workshop?

Contact us!

Additional Technical Details

MGI Schema Repository and Registry: This service is operated to improve discovery, access and reuse of materials-related schemas, data models, ontologies, etc. https://schemas.nist.gov/