We are building a Data Commons for the biomaterial community that will enable users to compare biomaterial formulations with commercial materials in terms of their performance. Our aim is to accelerate biomaterials R&D for everyone, everywhere.

Data modelling and visualization

We are building an open repository of material property and performance data that can be used for material analyses, digital modelling and product design. This will include mechanical, thermal, barrier and aesthetic properties for our recipes. Machine learning algorithms will be integrated to enable novel formulation prediction from existing data and suggest candidate materials for fabrication, characterization, and model improvement. View our sample dataset of seaweed bioplastics.

Data modeling is being developed in collaboration with with the Department of Mechanical Engineering, University of Oxford through the EPSRC-funded project Poly(ML): Machine Learning for Improved Sustainable Plastics.

Data generation

We are combining data mining with distributed testing to generate data. We plan to develop open source testing devices with the international Fab Labs network. This opens up the potential for local materials testing and distributed data generation.

Open source hardware for materials characterization is being developed in collaboration with the Center for Bits and Atoms (CBA), MIT, Argonne National Labs, and the National Institute for Standards and Technology on the project OM3: Open Materials Metrology and Modeling.

Data Repository Implementation

The repository infrastructure is being built in accordance with FAIR Digital Object Framework, which will help enable the FAIR Data Principles of Findable, Accessible, Interoperable, and Reusable. Our platform will implement these principles through the use of Cordra and an extended implementation of schema.org. The Digital Object Architecture approach was recently highlighted in an RDA Adoption Story, and extended versions of schema.org are gaining adoption in many communities, such as life sciences with Bioschemas. The platform will enable Linked Data, or Semantic Web technology allowing data to be shared and reused between applications, organizations, and communities. Data from numerous distributed producers will be referenced with unique, persistent identifiers and a public-facing API will enable the community to develop apps for reading and writing data. Our use of schema.org/DefinedTerm will enable agile development in response to user needs as we contribute to an emerging vocabulary for biomaterials across communities of research and practice.

Data infrastructure is being developed in collaboration with the Materials Genome Initiative, National Institute of Standards and Technology (NIST).