Enzymes' characterization via spectral analysis of the Laplacian associated to their relative contact maps

The main motivation for my thesis is the believe that global properties of enzymes areessential for a complete understanding of their behaviors. In my thesis, in particular,I investigate qualitative properties of enzymes via spectral techniques associated to thegraph Laplacian. I try to apply visualization techniques to understand similarities anddissimilarities among different enzymes’ structures, encoded in adjacency matrices re-trieved from coordinate data in online available datasets. The purpose is to make anexploration of features and see whether these techniques, that are used extensively inliterature for visual discrimination tasks, are also useful for these biological entities.I have tried to design a size-independent analysis that would be able to differentiateamong different taxonomies, different catalytic properties and different environmentsassociated to enzymes. This attempt provided useful hints for the analysis of enzymeproperties, even if as a final remark the dependence from enzyme size is still found inthe Laplacian eigenvalue spectrum.