Method for predicting membrane protein beta-barrel transmembrane area based on sparse coding and chain training

Abstract

The invention provides a method for predicting a membrane protein beta-barrel transmembrane area based on sparse coding and chain training and relates to a sparse coding technology, a chain learning algorithm and a support vector machine. Structure prediction is conducted on the membrane protein beta-barrel transmembrane area through a computing method, and important information is provided for the research of the structures and functions of proteins. According to the method, the concept of digital image processing is introduced creatively, sparse coding is conducted on a protein feature matrix, and feature dimensionality reduction and denoising are achieved; a membrane protein beta-barrel data set is organized in a protein database PDB, a position-specific scoring matrix and a Z score are extracted and used as features, the position-specific scoring matrix represents amino acid evolution information, the Z score represents the position information of amino acid residues, a feature vector is extracted through a sliding window, multi-feature fusion is achieved, a chain learning algorithm training model based on a SVM classifier is provided, a predication effect is remarkably improved, and a Jakenife cross validation result shows that the precision can reach 92.5%.

Description

technical field [0001] The invention relates to a technology in the field of membrane protein structure prediction and computational intelligence, specifically a method for predicting the transmembrane region of membrane protein beta-barrel based on sparse coding and chain learning. Background technique [0002] At present, with the rapid development of proteome databases, the number of proteins with known structures is increasing, which plays an important role in the study of protein functions. Membrane proteins are embedded in the biomembrane and run through the phospholipid bilayer. They have strong hydrophobicity and are not suitable for crystallization. The experimental method to solve the protein structure is not only expensive but also time-consuming. Therefore, using calculation methods to predict protein structure is a However, traditional machine learning methods still have some problems to be solved in the field of protein prediction, such as feature selection and...

AI Technical Summary

Problems solved by technology

At present, methods for obtaining protein structures are based on statistical information and methods based on physical and chemical properties of membrane proteins Freeman, T. and Wimley, W. (2010) A highly accurate statistical approach for the prediction of transmembrane beta-barrels. Bioinformatics, such Methods based on statistical information and physicochemical properties of membrane proteins are limited to a small number of structurally simple protein types, such as membrane protein structures with a small number of beta-strands. With the rapid development of machine learning methods, such as based on hidden Mark Singh, N. et al. (2011) Tmbhmm: a frequency profile based HMM for predicting the topology of transmembrane beta barrel proteins and the exposure status of transmembrane residues. Biochim. Biophys. Acta BBA Proteins Proteomics, 1814, 664 –670, the prediction accuracy has been improved, but for special lengths such as shorter strands, there is a phenomenon that the false positive rate is too high, and the influence of system noise and many factors that reduce the prediction accuracy in the feature extraction process need to be resolved.

Images