pymatgen Materials Genomics ToolkitNew

What it is

pymatgen (Ong et al., Comput. Mater. Sci. 2013) is the de-facto Python library for materials analysis. It sits between raw DFT outputs (VASP, Quantum ESPRESSO, CP2K) and your analysis or ML pipeline.

Core use cases

• Structure manipulation: symmetry, surfaces, slabs, adsorption sites
• Phase diagrams: construction and stability analysis
• Electronic structure: band structures, DOS, Fermi surfaces
• Interfaces: Materials Project REST API and the legacy data layer

Learning path

1. Parse a POSCAR/CIF; explore symmetry and neighbors
2. Build a phase diagram from MP entries; identify stable phases
3. Slab construction and adsorption-site enumeration for catalysis studies
4. Convert structures into ML inputs (graphs, descriptors) for GNN or MLIP work

Why it matters for physics-AI

Almost every public materials-ML dataset is consumed and analyzed through pymatgen objects - learning it early removes friction everywhere downstream.