Hierarchical Description of Protein Structure Fragments obtained from Analyses of Promolecular Electron Density Distributions

Various macromolecular modelling programs are based on a dictionnary of protein fragments. For example, electron density (ED) map interpretation programs, such as O, warpNtrace, MAIN, ... use short segments such as secondary structure templates, side-chain conformers, ... to be fitted, either manually or automatically, in an ED map. These fragments are determined from a statistical analysis of PDB structures, users own data, ... In this abstract, a theoretical method applied to describe molecular structures (small or large molecules) in terms of hierarchically related substructures is presented [1]. The approach is based on the location of local maxima (peaks) in promolecular ED distributions established at continuously varying smoothing degrees t. The local maxima are determined using a hierarchical clustering algorithm [2] wherein peaks obtained at a given resolution are used as starting points for discovering peaks at the next lower resolution level through gradient trajectories of the ED distribution. The use of such an approach allows to assign molecular fragments or chemical groups to peaks, at any smoothing level. Promolecular ED distributions are analytically represented using either the Promolecular Atom Shell Approximation (PASA) [3] or the Cromer-Mann coefficients as available in XTAL [4], using relationships such as: The analysis of the decomposition patterns of a small molecules and protein structures are presented. Applications are suggested in pharmacophore definition and protein geometry showed that the amino-acid residues have a similar decomposition pattern regardless of their position in the protein sequence, the protein conformation, and the influence of the crystal packing. [1] Leherte, L., Dury, L. & Vercauteren, D. P. (2003). J. Phys. Chem. A 107, 9875-9886. [2] Leung, Y., Zhang, J.-S. & Xu, Z.-B. (2000). IEEE Trans. on Pattern Analysis and Machine Intelligence 22, 1396-1410. [3] Amat, L., Carbó-Dorca, R. (1997). J. Comput. Chem. 19, 2023-2039; Coefficients and exponents can be seen and downloaded from the Web site: http://iqc.udg.es/cat/similarity/ASA/funcset.html. [4] Hall, S., du Boulay, D. & Olthof-Hazekamp, R., Eds. (2002). The Gnu Xtal System of Crystallographic Software, v. 3.7.2; http://xtal.sourceforge.net/.