PyMEDA: A unified python framework for multiscale energy decomposition Analysis

Energy decomposition analysis (EDA) has become a powerful approach for elucidating chemical bonding and non-covalent interactions, especially in complex and multiscale systems. Embedding strategies combining high-level quantum chemical methods with more approximate treatments of the environment provide a promising balance between accuracy and efficiency, yet existing embedding schemes of EDA approaches implementations are often fragmented and software-dependent. To address this challenge, we present PyMEDA, a unified, open-source interface framework for multiscale energy decomposition analysis. Through a modular design, PyMEDA standardizes interfaces to diverse quantum mechanical, molecular mechanics, and semi-empirical backends (initially XEDA, OpenMM, and xtb), enabling the unified implementation of workflows for both conventional and embedding-type EDA schemes, such as DM-EDA (QM/MM) and DM-EDA (EB). By providing extensibility, reproducibility, and automation, PyMEDA offers a flexible environment to advance multiscale interaction analysis and facilitates the future development of embedding-based EDA methods.