OCLSTM: Optimized convolutional and long short-term memory neural network model for protein secondary structure prediction.

Alexandre G. de Brevern,Yihui Liu,Yawu Zhao

doi:10.1371/journal.pone.0245982

Alexandre G. de Brevern, Yihui Liu + Show 1 more

Open Access

https://doi.org/10.1371/journal.pone.0245982

Copy DOI

Abstract

Protein secondary structure prediction is extremely important for determining the spatial structure and function of proteins. In this paper, we apply an optimized convolutional neural network and long short-term memory neural network models to protein secondary structure prediction, which is called OCLSTM. We use an optimized convolutional neural network to extract local features between amino acid residues. Then use the bidirectional long short-term memory neural network to extract the remote interactions between the internal residues of the protein sequence to predict the protein structure. Experiments are performed on CASP10, CASP11, CASP12, CB513, and 25PDB datasets, and the good performance of 84.68%, 82.36%, 82.91%, 84.21% and 85.08% is achieved respectively. Experimental results show that the model can achieve better results.

Highlights

Protein is the material basis of life activities, the basic organic matter that constitutes cells, and the main bearer of life activities
In the field of bioinformatics, protein secondary structure prediction is a very important and challenging problem [1], it is difficult to predict the spatial structure of a protein from a primary structure, so the prediction of protein secondary structure has been valued by many people
BiLSTM neural network can capture the long-distance dependence of the amino acid features extracted by the convolutional neural network

Summary

Introduction

Protein is the material basis of life activities, the basic organic matter that constitutes cells, and the main bearer of life activities. The structure of a protein mainly includes primary structure, secondary structure, tertiary structure and quaternary structure. Protein secondary structure is formed by folding based on protein primary structure. The tertiary structure of the protein is further coiled and folded based on the secondary structure, and the specific spatial structure formed by the maintenance of the secondary bond is called the tertiary structure of the protein. Protein quaternary structure refers to a polymer structure formed by connecting multiple polypeptide chains with independent tertiary structure through non-covalent bonds. In the field of bioinformatics, protein secondary structure prediction is a very important and challenging problem [1], it is difficult to predict the spatial structure of a protein from a primary structure, so the prediction of protein secondary structure has been valued by many people

Methods

Results

Conclusion