Three-dimensional Krawtchouk descriptors for protein local surface shape comparison. (September 2019)

• Record Type: Journal Article
• Authors: Sit, Atilla; Shin, Woong-Hee; Kihara, Daisuke
• Is Part Of: Pattern recognition. Volume 93(2019:Sep.)
• Page Start: 534
• Page End: 545
• Publication Date: 2019-09
• Subjects: 3D image retrieval -- Local image comparison -- Region of interest -- Discrete orthogonal functions -- Krawtchouk polynomials -- Weighted Krawtchouk polynomials -- 3D Krawtchouk moments -- Protein surface -- Ligand binding site -- Pocket comparison -- Structure-based function prediction; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.05.019
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: 2D Krawtchouk descriptors have been extended to 3D for local comparison of 3D surfaces. The location of the region-of-interest can be controlled by changing three parameters. The new formulation has many advantages over similar approaches, including minimal redundancy and discriminative performance. The new method uses only a small number of invariant descriptors per image for an efficient local image retrieval. The new descriptors have the potential to be applied to predicting biological function of proteins and computational drug design. Abstract: Direct comparison of three-dimensional (3D) objects is computationally expensive due to the need for translation, rotation, and scaling of the objects to evaluate their similarity. In applications of 3D object comparison, often identifying specific local regions of objects is of particular interest. We have recently developed a set of 2D moment invariants based on discrete orthogonal Krawtchouk polynomials for comparison of local image patches. In this work, we extend them to 3D and construct 3D Krawtchouk descriptors (3DKDs) that are invariant under translation, rotation, and scaling. The new descriptors have the ability to extract local features of a 3D surface from any region-of-interest. This property enables comparison of two arbitrary local surface regions from different 3D objects. We present the new formulation of 3DKDs and apply it to the local shape comparison of protein surfaces in order to predict ligand molecules that bind to query proteins.