ABSTRACT
The secondary structure of protein has three-state and eight-state classification standards, and there exist certain correspondences and commonalities between the two. In order to make full use of the consistent relationship to improve protein secondary structure prediction, an ensemble method named TSAES is introduced, which enables the three-classification and eight-classification to correct each other to attain more reasonable. Based on the same considerations, the final Multi-Output Simultaneous Prediction deep model shorted for MOSP is further proposed. It first mines the feature scale of the protein sequence, and then leverages a multi-layer temporal convolution networks to effectively extract the short-range dependence and long-term information of the sequence. High-level feature representation is finally used to optimize the outputs of three-state and eight-state classifications simultaneously. Experimental results demonstrate that 38relation-aware methods proposed here averagely enhanced 0.73%, 0.48% for Q3 and Q8 accuracy upon 38relation-agnostic baseline model. Though our work, the effectiveness of relation utilization had been verified, hence a novel valuable thinking angle except model innovation when making accurate prediction have been provided. Moreover, the proposed methods could be easily transplanted to other models, exhibiting strong feasibility and practicality.
- Kabsch W, Sander C. Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features[J]. Biopolymers: Original Research on Biomolecules, 1983, 22(12): 2577-2637.Google Scholar
- McGuffin L J, Jones D T. Benchmarking secondary structure prediction for fold recognition[J]. Proteins: Structure, Function, and Bioinformatics, 2003, 52(2): 166-175.Google Scholar
- Zhou J, Troyanskaya O G. Deep supervised and convolutional generative stochastic network for protein secondary structure prediction[C]//31st International Conference on Machine Learning, ICML 2014. International Machine Learning Society (IMLS), 2014: 1121-1129.Google Scholar
- Kaae Sønderby S, Winther O. Protein Secondary Structure Prediction with Long Short Term Memory Networks[J]. arXiv preprint arXiv:1412.7828, 2014.Google Scholar
- Zheng L, Li H, Wu N, Protein secondary structure prediction based on deep learning[J]. DEStech Transactions on Engineering and Technology Research, 2017.Google ScholarCross Ref
- Li Z, Yu Y. Protein secondary structure prediction using cascaded convolutional and recurrent neural networks[C]//Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence. AAAI Press, 2016: 2560-2567.Google Scholar
- Wang S, Peng J, Ma J, Protein secondary structure prediction using deep convolutional neural fields[J]. Scientific reports, 2016, 6(1):1-11.Google Scholar
- Wang Y, Mao H, Yi Z. Protein secondary structure prediction by using deep learning method[J].Knowledge-Based Systems, 2017, 118:115-123.Google ScholarCross Ref
- Lea C, Flynn M D, Vidal R, Temporal convolutional networks for action segmentation and detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2017: 1003-1012.Google Scholar
- Bai S, Kolter J Z, Koltun V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[J]. arXiv preprint arXiv:1803.01271, 2018.Google Scholar
- Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[J]. arXiv preprint arXiv:1511.07122, 2015.Google Scholar
- He K, Zhang X, Ren S, Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.Google Scholar
- Gal Y, Ghahramani Z. A theoretically grounded application of dropout in recurrent neural networks[C]//Advances in neural information processing systems. 2016: 1019-1027.Google Scholar
- Abadi M, Agarwal A, Barham P, Tensorflow: Large-scale machine learning on heterogeneous distributed systems[J]. arXiv preprint arXiv:1603.04467, 2016.Google Scholar
- Kingma D P, Ba J. Adam: A method for stochastic optimization[C]// International Conference on Learning Representations (ICLR). 2015:1-13.Google Scholar
- Zhang C T, Zhang R. A refined accuracy index to evaluate algorithms of protein secondary structure prediction[J]. Proteins: Structure, Function, and Bioinformatics, 2001, 43(4): 520-522.Google ScholarCross Ref
- Zhang L, Li Z, Yang F B, Eight-category protein secondary prediction algorithm based on deep learning[J]. Computer Applications, 2017, 37(05):1512-1515.Google Scholar
- Johansen A R, Sønderby C K, Sønderby S K, Deep recurrent conditional random field network for protein secondary prediction[C]//Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics. ACM, 2017: 73-78.Google Scholar
- 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): 293-293.Google ScholarCross Ref
- Guo Yan-fu, Li Wei-hua, Wang Bing-yi, Protein secondary structure prediction based on Convolutional Long Short-term Memory Networks[J]. Pattem Recognition and Aitificial Intelligence, 2018, 31(06):562-568.Google Scholar
- Busia A, Collins J, Jaitly N. Protein secondary structure prediction using deep multi-scale convolutional neural networks and next-step conditioning[J]. arXiv preprint arXiv:1611.01503, 2016.Google Scholar
- Drori I, Dwivedi I, Shrestha P, High quality prediction of protein Q8 secondary structure by diverse neural network architectures[J]. arXiv preprint arXiv:1811.07143, 2018.Google Scholar
- Vaswani A, Shazeer N, Parmar N, Attention is all you need[C]//Advances in neural information processing systems. 2017:5998-6008.Google Scholar
Recommendations
Protein Secondary Structure Prediction based on Wavelets and 2D Convolutional Neural Network
CSBio '16: Proceedings of the 7th International Conference on Computational Systems-Biology and BioinformaticsIn this paper we propose an approach to use wavelets and 2D convolutional neural network (CNN) to extract features for the prediction of protein secondary structure. A wavelet feature matrix extracted from PSSM profiles is input into convolutional ...
Improving prediction of protein secondary structure using physicochemical properties of amino acids
ISB '10: Proceedings of the International Symposium on BiocomputingProtein Structure Prediction is important in the sense that it helps to extend knowledge about the understanding of protein structures and functions. The knowledge is essential for prediction of secondary structures of unknown proteins required for ...
Weave amino acid sequences for protein secondary structure prediction
DMKD '03: Proceedings of the 8th ACM SIGMOD workshop on Research issues in data mining and knowledge discoveryGiven a known protein sequence, predicting its secondary structure can help understand its three-dimensional (tertiary) structure, i.e., the folding. In this paper, we present an approach for predicting protein secondary structures. Different from the ...
Comments