Speaker
Description
The FAIRmat consortium develops an open, interoperable research data infrastructure for computational and experimental materials science, with a strong focus on implementing the FAIR principles across heterogeneous workflows. Within this context, NeXus serves as an important standard for structuring experimental data in the NOMAD platform.
We present how NeXus is integrated into NOMAD’s ingestion, normalization, and archival pipelines. Central is the mapping of NeXus application definitions (NXDL) to a unified metadata schema, enabling automated parsing of NeXus/HDF5 files, annotation using external domain ontologies, and enrichment with provenance metadata. We discuss challenges such as schema mismatches, evolving NXDL definitions, and heterogeneous experimental setups.
We further show how NeXus-based datasets are propagated through NOMAD workflows, supporting advanced search, cross-experiment comparison, and integration with computational data, while preserving provenance from acquisition to publication.
Finally, we outline future directions, including tighter coupling with semantic web technologies and integration of external knowledge bases (e.g., ontologies and reference databases) via persistent identifiers. This aims to enhance machine-actionable interoperability across platforms and disciplines.
