Method and device for predicting topological structure of α-helical transmembrane proteins

Abstract

A method for predicting the topology of α-helix transmembrane proteins, organizing training sets, validation sets and test sets according to the definition of transmembrane α-helix TMH; extracting position-specific scoring matrices PSSM and HMM from the sequences in the training set, validation set and test set , water solubility, secondary structure, torsion angle, and hydrophilic index features; use the training set to train the deep residual network model based on the entire sequence and the deep residual network model based on the sliding window. After integrating the average value of the output of the two networks, the dynamic threshold algorithm is used to obtain the TMH area; the training set is used to train the support vector machine model. The input to the model is the junction of other regions non‑TMH and TMH regions; the output is the position of non‑TMH relative to the cell membrane. Firstly predict the TMH region in the protein, then predict the position of non-TMH, and combine the prediction results of the two parts to get the final topology of the protein.

Description

technical field [0001] The invention belongs to the technical field of biological detection, in particular to a method and device for predicting the topological structure of α-helical transmembrane proteins based on multi-scale deep learning. Background technique [0002] The cell membrane is the barrier of the cell, which can isolate the internal environment of the cell from the external environment. The cell membrane is composed of a phospholipid bilayer and a large number of membrane proteins embedded in it. Membrane proteins play an important role in a series of biological processes such as cell signal transduction, ion conductivity, cell aggregation, cell recognition and intercellular communication. Therefore, many drugs are designed to bind to membrane proteins and thereby affect biological processes. [0003] Among all membrane proteins, α-helical transmembrane proteins account for the majority. It is estimated that 27% of the proteins in the human body are alpha-h...

AI Technical Summary

Problems solved by technology

In addition, as the evaluation criteria become more and more stringent, there is still room for improvement in the accuracy of previous prediction algorithms

Images