Secondary protein structureprediction method based on deep neural network

A deep neural network and secondary structure technology, applied in the fields of feature learning, deep learning, neural network, protein structure prediction and sequence learning. It can improve the reliability of prediction, reduce the difficulty of parameter selection, and improve the robustness of the system.

Inactive Publication Date: 2016-09-07
SICHUAN UNIV
View PDF2 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In view of the above technical problems, the present invention provides an efficient protein secondary structure prediction method, which can more accurately predict the secondary structure of amino acid residues in protein sequences; it aims to solve the problem that the existing technology cannot fully utilize the residue information between sequences , cannot meet the requirements of computing efficiency and accuracy for big data environment system structure prediction tasks, choose convergence speed and network parameters difficult and poor reliability and other technical problems

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Secondary protein structureprediction method based on deep neural network
  • Secondary protein structureprediction method based on deep neural network
  • Secondary protein structureprediction method based on deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] A protein secondary structure prediction method based on a deep neural network, comprising the following steps:

[0036] Step 1, the model training phase, which includes:

[0037] Obtain protein sequence combination features, position-specific scoring matrices (PSSM, position-specific scoring matrices), physical and chemical features as input, and train the self-encoder network to extract effective features;

[0038] Taking the protein sequence combination features of the independent training set as input and the corresponding secondary structure sequence as the target, the deep revertive neural network is trained by supervised learning to predict the secondary structure of each residue site.

[0039] Step 2, the prediction stage, which includes:

[0040] Input protein sequence features and predict the secondary structure of residues at each site.

[0041] In the above technical solution, the feature extraction in step 1 is from the encoder pre-training process, inclu...

Embodiment 2

[0064] see Figure 6 , a protein secondary structure prediction method based on a deep neural network, first, input a protein sequence feature combination sequence, which includes the combination features (PSSM, physicochemical features, etc.) of each residue site in the protein sequence. The input data needs to be preprocessed, which includes standardization, feature dimension alignment, etc. The final input is a protein sequence feature matrix.

[0065] The model training stage is to train the secondary structure prediction model. The specific process is as follows:

[0066] 1) Pre-trained self-encoding feature extraction network. A layer-by-layer training algorithm based on BP algorithm is used to train multiple autoencoders and stacked connections to form an autoencoder network. The network is a multi-layer deep neural network, such as figure 1 shown. The pre-trained self-encoder network is used for the preliminary feature extraction of the input protein combination f...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a secondary protein structureprediction method based on a deep learning andneural network method, and relates to the technical field of neural networks and secondary protein structureprediction. The method comprises the steps of inputting a protein characteristic sequence, and predicting a space secondary structure of an amino acid residue at each site of the sequence through a designed deep recurrent neural network model. The method provided by the invention realizes automatic predication of the secondary structure based on input characteristics, has better generalization ability, and can train a specific model and realize secondary structure predication with high accuracy according to the different input characteristics.

Description

[0001] technology neighborhood [0002] The invention relates to the fields of feature learning, neural network, deep learning, protein structure prediction and sequence learning, and in particular to a protein secondary structure prediction method based on deep neural network. Background technique [0003] The protein structure prediction problem is one of the important research problems in computational biology, which can discover the complex relationship between protein sequence structure and its function, and the secondary structure prediction problem is the basis of various higher-level structure prediction problems. Through accurate protein secondary structure prediction, researchers can quickly obtain the secondary structure conformation information of amino acid residues in the protein sequence, such as α-helix, β-sheet and irregular coil, etc. Analytics provide effective data reference and are widely adopted. [0004] Determining protein structure through experiments...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F19/16G06N3/08
CPCG06N3/08G16B15/00
Inventor 毛华陈媛媛罗川汪洋旭陈盈科
Owner SICHUAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap