Deep Learning Framework for Predicting Essential Proteins with Temporal Convolutional Networks

doi:10.1007/s12204-023-2632-9

Abstract

Abstract: Essential proteins are an indispensable part of cells and play an extremely significant role in genetic disease diagnosis and drug development. Therefore, the prediction of essential proteins has received extensive attention from researchers. Many centrality methods and machine learning algorithms have been proposed to predict essential proteins. Nevertheless, the topological characteristics learned by the centrality method are not comprehensive enough, resulting in low accuracy. In addition, machine learning algorithms need sufficient prior knowledge to select features, and the ability to solve imbalanced classification problems needs to be further strengthened. These two factors greatly affect the performance of predicting essential proteins. In this paper, we propose a deep learning framework based on temporal convolutional networks to predict essential proteins by integrating gene expression data and protein-protein interaction (PPI) network. We make use of the method of network embedding to automatically learn more abundant features of proteins in the PPI network. For gene expression data, we treat it as sequence data, and use temporal convolutional networks to extract sequence features. Finally, the two types of features are integrated and put into the multi-layer neural network to complete the final classification task. The performance of our method is evaluated by comparing with seven centrality methods, six machine learning algorithms, and two deep learning models. The results of the experiment show that our method is more effective than the comparison methods for predicting essential proteins.

Key words: temporal convolutional networks, node2vec, protein-protein interaction (PPI) network, essential proteins, gene expression data

摘要： 关键蛋白质是细胞不可缺少的组成部分，并且在遗传病诊断和药物开发中发挥着极其重要的作用。因此，关键蛋白质的预测受到了研究人员的广泛关注。许多中心性方法和机器学习算法已经被提出来预测关键蛋白质。然而，中心性方法学习到的拓扑特征不够全面，导致准确率较低。此外，机器学习算法需要足够的先验知识来选择特征，解决不平衡分类问题的能力有待进一步加强。这两个因素极大地影响了预测关键蛋白质的性能。在本文中，我们提出了一种基于时间卷积网络的深度学习框架，通过整合基因表达数据和蛋白质相互作用（PPI）网络来预测关键蛋白质。我们利用网络嵌入的方法自动学习PPI网络中蛋白质更丰富的特征。对于基因表达数据，我们将其视为序列数据，并使用时间卷积网络提取序列特征。最后将两类特征融合在一起，放入多层神经网络中，完成最终的分类任务。通过与7种中心性方法、6种机器学习算法和2种深度学习模型进行比较，评估了我们方法的性能。实验结果表明，我们的方法比预测关键蛋白质的比较方法更有效。

关键词: 时间卷积网络，node2vec，蛋白质相互作用网络，关键蛋白质，基因表达数据

CLC Number:

TP393
Q51

Lu Pengli, Yang Peishi, Liao Yonggang. Deep Learning Framework for Predicting Essential Proteins with Temporal Convolutional Networks[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(3): 510-520.

References

[1] CLATWORTHY A E, PIERSON E, HUNG D T. Targeting virulence: A new paradigm for antimicrobial therapy [J]. Nature Chemical Biology, 2007, 3(9): 541- 548.

[2] GIAEVER G, CHU A M, NI L, et al. Functional profiling of the saccharomyces cerevisiae genome [J]. Nature, 2002, 418(6896): 387-391.

[3] CULLEN L M, ARNDT G M. Genome-wide screening for gene function using RNAi in mammalian cells [J]. Immunology and Cell Biology, 2005, 83(3): 217-223.

[4] ROEMER T, JIANG B, DAVISON J, et al. Large-scale essential gene identification in Candida albicans and applications to antifungal drug discovery [J]. Molecular Microbiology, 2003, 50(1): 167-181.

[5] FREEMAN L C. Centrality in social networks conceptual clarification [J]. Social Networks, 1978, 1(3): 215-239.

[6] JOY M P, BROCK A, INGBER D E, et al. Highbetweenness proteins in the yeast protein interaction network [J]. Journal of Biomedicine and Biotechnology, 2005, 2005(2): 96-103.

[7] WUCHTY S, STADLER P F. Centers of complex networks [J]. Journal of Theoretical Biology, 2003, 223(1): 45-53.

[8] ESTRADA E, RODR´IGUEZ-VEL´AZQUEZ J A. Subgraph centrality in complex networks [J]. Physical Review E, 2005, 71(5): 056103.

[9] WANG J X, LI M, WANG H, et al. Identification of essential proteins based on edge clustering coefficient [J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2012, 9(4): 1070-1080.

[10] LI M, WANG J X, CHEN X, et al. A local average connectivity-based method for identifying essential proteins from the network level [J]. Computational Biology and Chemistry, 2011, 35(3): 143-150.

[11] ZHANG C L, ZHANG S W. A supervised orthogonal discriminant projection for tumor classification using gene expression data [J]. Computers in Biology and Medicine, 2013, 43(5): 568-575.

[12] WANG J S, WANG N, GE F, et al. Gene expression data classification using Laplacian eigenmap based on improved maximum margin criterion [J]. Chinese Journal of Electronics, 2013, 22(3): 521-524.

[13] TANG X W, WANG J X, ZHONG J C, et al. Predicting essential proteins based on weighted degree centrality [J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2014, 11(2): 407- 418.

[14] YUGANDHAR K, GROMIHA M M. Feature selection and classification of protein-protein complexes based on their binding affinities using machine learning approaches [J]. Proteins: Structure, Function, and Bioinformatics, 2014, 82(9): 2088-2096.

[15] LUO J W, QI Y. Identification of essential proteins based on a new combination of local interaction density and protein complexes [J]. PLoS One, 2015, 10(6): e0131418.

[16] LI M, LU Y, NIU Z B, et al. United complex centrality for identification of essential proteins from PPI networks [J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2017, 14(2): 370-380.

[17] LI M, ZHANG H H,WANG J X, et al. A new essential protein discovery method based on the integration of protein-protein interaction and gene expression data [J]. BMC Systems Biology, 2012, 6(1): 1-9.

[18] XIAO Q H, WANG J X, PENG X Q, et al. Identifying essential proteins from active PPI networks constructed with dynamic gene expression [J]. BMC Genomics, 2015, 16(Suppl 3): S1.

[19] HWANG Y C, LIN C C, CHANG J Y, et al. Predicting essential genes based on network and sequence analysis [J]. Molecular BioSystems, 2009, 5(12): 1672-1678.

[20] ACENCIO M L, LEMKE N. Towards the prediction of essential genes by integration of network topology, cellular localization and biological process information [J]. BMC Bioinformatics, 2009, 10(1): 1-18.

[21] ZENG M, LI M, WU F X, et al. DeepEP: A deep learning framework for identifying essential proteins [J]. BMC Bioinformatics, 2019, 20(Suppl 16): 506.

[22] ZHANG X E, XIAOW X, XIAOW J. DeepHE: Accurately predicting human essential genes based on deep learning [J]. PLoS Computational Biology, 2020, 16(9): e1008229.

[23] GROVER A, LESKOVEC J. Node2vec: Scalable feature learning for networks [C]//22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco: ACM, 2016: 855-864.

[24] BAI S J, KOLTER J Z, KOLTUN V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling [DB/OL]. (2018-03-04). https://arxiv.org/abs/1803.01271

[25] NAIR V, HINTON G E. Rectified linear units improve restricted boltzmann machines Vinod Nair [C]//27th International Conference on Machine Learning. Haifa: ICML, 2010: 807-814.

[26] IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift [C]//32nd International Conference on International Conference on Machine Learning. New York: ACM, 2015: 448-456.

[27] ZAHIDI Y, EL YOUNOUSSI Y, AZROUMAHLI C. Comparative study of the most useful Arabicsupporting natural language processing and deep learning libraries [C]//2019 5th International Conference on Optimization and Applications. Kenitra: IEEE, 2019: 1-10.

[28] BUNDY A, WALLEN L. Breadth-first search [M]//Catalogue of artificial intelligence tools. Berlin, Heidelberg: Springer, 1984: 13.

[29] TARJAN R. Depth-first search and linear graph algorithms [J]. SIAM Journal on Computing, 1972, 1(2): 146-160.

[30] CHATR-ARYAMONTRI A, OUGHTRED R, BOUCHER L, et al. The BioGRID interaction database: 2017 update [J]. Nucleic Acids Research, 2017, 45(D1): D369-D379.

[31] MEWES H W, FRISHMAN D, G¨ULDENR U, et al. MIPS: a database for genomes and protein sequences [J]. Nucleic Acids Research, 2002, 30(1): 31-34.

[32] CHERRY J M, ADLER C, BALL C, et al. SGD: Saccharomyces genome database [J]. Nucleic Acids Research, 1998, 26(1): 73-79.

[33] ZHANG R, LIN Y. DEG 5.0, a database of essential genes in both prokaryotes and eukaryotes [J]. Nucleic Acids Research, 2009, 37(suppl 1): D455-D458.

[34] WINZELER E A, SHOEMAKER D D, ASTROMOFF A, et al. Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis [J]. Science, 1999, 285(5429): 901-906.

[35] TU B P, KUDLICKI A, ROWICKA M, et al. Logic of the yeast metabolic cycle: Temporal compartmentalization of cellular processes [J]. Science, 2005, 310(5751): 1152-1158.

[36] PEDREGOSA F, VAROQUAUX G, GRAMFORT A, et al. Scikit-learn: Machine learning in python [DB/OL]. (2012-01-02). https://arxiv.org/ abs/1201.0490

[37] ZENG M, LI M, FEI Z H, et al. A deep learning framework for identifying essential proteins by integrating multiple types of biological information [J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2021, 18(1): 296-305.