FAIRmat Tutorial: Molecular Dynamics Trajectories and Workflows in NOMAD
Wednesday, 15. February 2023 // 14.00 - 17.00
Consortium FAIRmat, FAIR-DI e.V.
FAIRmat hands-on tutorial series
These tutorials offer current and potential users and members of the community a chance to learn about our work towards a FAIR data infrastructure for condensed-matter physics and the chemical physics of solids. We use real, hands-on examples to show how FAIRmat can help researchers today as well as to inform about upcoming new scientific opportunities.
Tutorial 7: Molecular Dynamics Trajectories and Workflows in NOMAD
The FAIRmat consortium is committed to extending the NOMAD infrastructure to a wide variety of materials science data. To support soft matter simulations (e.g., atomistic molecular dynamics simulations), a number of challenges arise, primarily due to the volume and variety of data.
The FAIRmat team is working to overcome these challenges, and the NOMAD infrastructure is now equipped with new metadata, features, and tools specifically designed to ease the FAIR treatment of trajectory data and workflows. Parsers have been implemented for two of the most popular molecular dynamics codes (Gromacs and Lammps), with plans for quick expansion to additional codes within the next year. The NOMAD Metainfo now describes the system’s hierarchical structure (in terms of bond topology) through the concept of fixed chemical bonds defined within classical force fields. The NOMAD GUI provides a bespoke overview page for molecular dynamics data, which includes tools that ease visualization of the system topology and automatically displays structural, dynamic, and thermodynamic observables that can assist in a fast assessment of system equilibration. Additionally, a native workflow visualizer allows the user to connect individual simulation entries into complex workflows. Finally, the NOMAD Python module facilitates custom trajectory analysis, for instance in a Jupyter notebook, with functions that convert a NOMAD archive entry to an instance of the MDAnalysis data class.
This tutorial invites both experienced and completely novice NOMAD users to learn about these new features for molecular dynamics trajectories. A brief introduction to the FAIRmat consortium and the NOMAD infrastructure will be given, followed by guided and interactive tutorials highlighting the various features described above.