Protein Secondary Structure Prediction based on Multi-scale Convolution and Transformer

Shiwei   Yang; Xiaozhou Chen

doi:10.54097/hf4n5726

Authors

Shiwei Yang
Xiaozhou Chen

DOI:

https://doi.org/10.54097/hf4n5726

Keywords:

Multi-scale Convolution, Structure Prediction, Transformer, Neural Network

Abstract

As an important field of natural science, bioinformatics reveals various complex biological facts. The task of protein structure prediction has always been an important research topic in bioinformatics, starting from the primary structure, namely the amino acid sequence, to predict the secondary structure of proteins. In this experiment, amino acid sequences were encoded in various ways, and the encoded data were input into the constructed deep learning network to predict the secondary structure of proteins. In this paper, an MCN-T network is constructed by combining CNN and Transformer. The powerful global feature capture and parallel processing capabilities of the Transformer are combined with the efficient local feature extraction and dimension reduction capabilities of CNN. This combination enables the model to perform well in protein secondary structure prediction tasks. In the experiment, the optimal form of the model is obtained after various parameter optimization experiments. The accuracy of the MCN-T network on the CB6133_filtered dataset and the CB513 dataset is 71.93% and 67.83%, respectively.

Downloads

Download data is not yet available.

References

Li, Zhen, and Yizhou Yu. "Protein secondary structure prediction using cascaded convolutional and recurrent neural networks." arxiv preprint arxiv:1604.07176 (2016).

Zhang B, Li J, Lü Q. Prediction of 8-state protein secondary structures by a novel deep learning architecture[J]. BMC bioinformatics, 2018, 19: 1-13.

Guo Y, Wang B, Li W, et al. Protein secondary structure prediction improved by recurrent neural networks integrated with two-dimensional convolutional neural networks[J]. Journal of bioinformatics and computational biology, 2018, 16(05): 1850021.

Hu H, Li Z, Elofsson A, et al. A Bi-LSTM based ensemble algorithm for prediction of protein secondary structure[J]. Applied Sciences, 2019, 9(17): 3538.

Zhao Y, Zhang H, Liu Y. Protein secondary structure prediction based on generative confrontation and convolutional neural network[J]. IEEE Access, 2020, 8: 199171-199178.

Zhao Yawu, Liu Yihui. Protein secondary structure prediction based on optimized convolutional neural networks [J]. Computer Applications and Software, 2021, 38(7): 147-152,166.

Altschul S F, Gertz E M, Agarwala R, et al. PSI-BLAST pseudocounts and the minimum description length principle[J]. Nucleic acids research, 2009, 37(3): 815-824.

Creighton T E. Proteins: structures and molecular properties[M]. Macmillan, 1993.

Hsu C W, Lin C J. A comparison of methods for multiclass support vector machines[J]. IEEE transactions on Neural Networks, 2002, 13(2): 415-425.

Wu Yuming. A new coding method for protein secondary structure prediction [J]. Industrial Control Computer, 2015 (4): 109-110.

Majid Vafaeipour, Omid Rahbari, Marc A. Rosen,Farivar Fazelpour,Pooyandeh Ansarirad. Application of sliding window technique for prediction of wind velocity time series[J]. International Journal of Energy and Environmental Engineering, 2014,5(2-3).

Zhao Y, Liu Y. OCLSTM: Optimized convolutional and long short-term memory neural network model for protein secondary structure prediction[J]. PLOS ONE, 2021, 16.

Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. Advances in neural information processing systems[J]. Advances in neural information processing systems, 2017, 30(2017).

Jiang Q, Jin X, Lee S J, et al. Protein Secondary Structure Prediction: A Survey of the state of the art [J]. Journal of molecular graphics & modelling, 2017, 76:379-402.

Zhang Tian-Ze, Li Yuan-Xiang, Xiang Zheng-Long, et al. Particle swarm optimization algorithm based on RMSprop [J]. Computer Engineering and Design, 2021, 42(3): 7.