ProDy is a Python package for structure-based analysis of protein dynamics. It allows for quantitative characterization of structural variations in heterogeneous datasets of experimentally resolved structures and comparison with theoretically predicted equilibrium dynamics. Datasets include structural ensembles for protein families, their mutants, sequence homologues, and structures in the presence/absence of substrates, ligands, or inhibitors. ProDy provides helper functions for comparative analysis of experimental and theoretical data and visualization of principal conformational changes. The API allows users to extend the software and implement new methods. ProDy is open source and freely available under the GNU General Public License. It can analyze PDB files or sequences and retrieve structures from the PDB. It also supports analysis of molecular dynamics (MD) trajectories. Experimental data include X-ray and NMR structures, which are often heterogeneous. ProDy implements algorithms for optimal alignment of such datasets and building covariance matrices. Principal component analysis (PCA) of the covariance matrix determines principal modes of structural variation. ProDy also supports Gaussian network model (GNM) and normal mode analysis (NMA) for theoretical dynamics. It offers essential dynamics analysis (EDA) of MD snapshots. ProDy provides a graphical interface, NMWiz, for ANM and PCA calculations within molecular visualization programs. It includes functions for structural alignment, comparison of multiple chains, and analysis of conformers. ProDy complements existing tools by allowing systematic retrieval and analysis of heterogeneous experimental datasets. It enables quantitative comparison of experimental datasets with theoretically predicted conformational dynamics. ProDy is extensible, interoperable, and provides scripts for automated tasks and batch analyses. It allows production of publication-quality figures and offers user-defined distance-dependent force functions. ProDy is a versatile, powerful tool for inferring protein dynamics from experiments and theory.ProDy is a Python package for structure-based analysis of protein dynamics. It allows for quantitative characterization of structural variations in heterogeneous datasets of experimentally resolved structures and comparison with theoretically predicted equilibrium dynamics. Datasets include structural ensembles for protein families, their mutants, sequence homologues, and structures in the presence/absence of substrates, ligands, or inhibitors. ProDy provides helper functions for comparative analysis of experimental and theoretical data and visualization of principal conformational changes. The API allows users to extend the software and implement new methods. ProDy is open source and freely available under the GNU General Public License. It can analyze PDB files or sequences and retrieve structures from the PDB. It also supports analysis of molecular dynamics (MD) trajectories. Experimental data include X-ray and NMR structures, which are often heterogeneous. ProDy implements algorithms for optimal alignment of such datasets and building covariance matrices. Principal component analysis (PCA) of the covariance matrix determines principal modes of structural variation. ProDy also supports Gaussian network model (GNM) and normal mode analysis (NMA) for theoretical dynamics. It offers essential dynamics analysis (EDA) of MD snapshots. ProDy provides a graphical interface, NMWiz, for ANM and PCA calculations within molecular visualization programs. It includes functions for structural alignment, comparison of multiple chains, and analysis of conformers. ProDy complements existing tools by allowing systematic retrieval and analysis of heterogeneous experimental datasets. It enables quantitative comparison of experimental datasets with theoretically predicted conformational dynamics. ProDy is extensible, interoperable, and provides scripts for automated tasks and batch analyses. It allows production of publication-quality figures and offers user-defined distance-dependent force functions. ProDy is a versatile, powerful tool for inferring protein dynamics from experiments and theory.