Protein data feature extraction method

Abstract

The invention discloses a protein data feature extraction method. The method comprises the following steps: (1) preprocessing an original three-dimensional model of protein to obtain a preprocessed three-dimensional model; (2) acquiring a plurality of two-dimensional views of the preprocessed three-dimensional model, extracting an image feature matrix of each two-dimensional view, and fusing all the image feature matrixes to obtain a two-dimensional feature matrix of the protein; (3) acquiring a three-dimensional characteristic matrix of the protein; and (4) performing fusion calculation on the two-dimensional feature matrix and the three-dimensional feature matrix of the protein to obtain a protein data feature matrix. According to the method, the two-dimensional view feature informationand the three-dimensional model space structure information of the protein are extracted, so that the appearance feature description of the protein is more comprehensive. Incompleteness caused by onlyadopting two-dimensional feature information extraction is avoided, and scientificity and accuracy of protein model similarity calculation can be guaranteed.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to a protein data feature extraction method. Background technique [0002] The database of protein molecules is rapidly increasing, and since proteins display multiple possible conformations in solution, the detection of shape similarity and identity is of biological relevance in the drug discovery process and in the molecular characterization of disease. Therefore, learning how to represent and highlight the features of protein 3D models is of great significance to both the medical field and the biological field. [0003] Currently, protein feature extraction methods are mainly divided into two types, one is protein feature extraction method based on 3D stereo model, and the other is protein feature extraction method based on multi-view. [0004] In protein feature extraction methods based on 3D volumetric models, proteins are described as model-based features,...

AI Technical Summary

Problems solved by technology

[0006] The present invention aims at the technical problems that the protein feature extraction method in the prior art only uses a three-dimensional model to extract a large amount of calculation, the accuracy is not high when two-dimensional feature extraction is used, and spatial information is easily lost, and a protein data feature extraction method is proposed. can solve the above problems

Images