Pattern recognition strategies for molecular surfaces. I. Pattern generation using fuzzy set theory

Abstract

A new method for the characterization of molecules based on the model approach of molecular surfaces is presented. We use the topographical properties of the surface as well as the electrostatic potential, the local lipophilicity/hydrophilicity, and the hydrogen bond density on the surface for characterization. The definition and the calculation method for these properties are reviewed shortly. The surface is segmented into overlapping patches with similar molecular properties. These patches can be used to represent the characteristic local features of the molecule in a way that is beyond the atomistic resolution but can nevertheless be applied for the analysis of partial similarities of different molecules as well as for the identification of molecular complementarity in a very general sense. The patch representation can be used for different applications, which will be demonstrated in subsequent articles.

Pattern recognition strategies for molecular surfaces: III. Binding site prediction with a neural network

“… Abstract An algorithm for the identification of possible binding sites of biomolecules, which are represented as regions of the molecular surface, is introduced. The algorithm is based on the segmentation of the molecular surface into overlapping patches as described in the first article of this series 1. The properties of these patches (calculated on the basis of physical and chemical properties) are used for the analysis of the molecular surfaces of 7821 proteins and protein complexes.…”

Pattern recognition strategies for molecular surfaces. II. Surface complementarity

“…1 The molecular representation as surface domains was used to identify binding sites in biological complexes and was implemented into a docking algorithm based on a fuzzy comparison using the 3D structure data of the compounds only. Even if the algorithm is not yet completely optimized and is not fully competitive in this current state to established docking algorithms like DOCK [2][3][4][5][6] or FlexX [7][8][9][10] in accuracy and speed, we think that the results are very promising for the identification of molecular complementarity in very large molecular systems and can contribute to the progress towards the automated identification of active sites in a wide application field.…”

Visual Analysis of Biomolecular Surfaces

“…Several approaches have been proposed to combine these two curvatures into a single value to describe local surface features [56,57,58,59,60].…”

“…Given a function f : M → R, one family of surface segmentation methods is based on the region-growing idea [56,65,59]. In particular, certain seeds are first selected, and neighboring points are gradually clustered into the regions around these seeds.…”