Methods of statistical and computational geometry, Voronoi and Delaunay tessellation in particular, play an increasingly important role in exploration of complex nature of molecular and biomolecular structure. The range of applications of statistical geometry for biomolecular structural studies has grown significantly in the past decade. As the new experimental structural information on biomolecules becomes available, the need for sophisticated and robust tools for its interpretation will further increase. At the same time more known structure will enable the creation of larger and better training sets for pattern identification. Existing and new statistical geometry algorithms may prove instrumental in future developments of methods for protein structure analysis, ab initio structure prediction, and protein engineering.