Cryogenic electron tomography (cryo-ET) allows studying biological macromolecular complexes in cells by threedimensional (3D) data analysis. The complexes continuously change their shapes (conformations) to achieve biological functions. The shape heterogeneity in cryo-ET is a bottleneck for comprehending biological mechanisms and developing drugs. Cryo-ET data suffer from a low signal-to-noise ratio and spatial anisotropies (missing wedge artefacts), making it particularly challenging for resolving the shape variability. Other shape variability analysis techniques simplify the problem by considering discrete rather than continuous conformational changes of complexes. Recently, HEMNMA-3D was introduced for cryo-ET continuous shape variability analysis, based on elastic and rigid-body 3D registration between simulated shapes and cryo-ET data using normal mode analysis and fast rotational matching with missing wedge compensation. HEMNMA-3D provides a visual insight into molecular dynamics by grouping and averaging subtomograms of similar shapes and by animating movies of registered motions. This article reviews HEMNMA-3D and compares it with existing literature on a simulated dataset for nucleosome shape variability.