Atomistic Schemas

The atomistic schema in atomea is designed to provide a flexible, rigorous, and theory-driven interface for interacting with data originating from simulations and calculations on atomic systems. In contrast to many domain-specific formats or narrowly scoped schemas (like those focused only on quantum chemistry, proteins, or molecular mechanics), atomea takes a pedantic and foundational approach. Our schemas describe data by its theoretical or mathematical meaning, decoupled from the idiosyncrasies of file formats, domain assumptions, or tool-specific naming conventions.

This design provides interoperability, clarity, and precision, making it easier to build robust tools, standardize datasets, and bridge communities working across different subfields.

Schema Design

Each ensemble is represented by an EnsembleSchema, which includes:

• identification: metadata and fingerprinting;
• system: atomic composition, charges, coordinates;
• topology: atom types, bonding, and groupings;
• energy: electronic, kinetic, and classical energies;
• qc: quantum chemistry parameters and results;
• time: time step and sampling intervals.

These fields are built from modular Pydantic models like SystemSchema, EnergySchema, QCSchema, and TimeSchema, all adhering to the same UUID/cadence principles.

Together, they form a rigorously structured yet highly extensible data model for representing atomic systems at scale.