Protein function prediction based on doubly indexed matrix

doi:10.11772/j.issn.1001-9081.2015.06.1637

Abstract

Abstract:

The single data source cannot effectively predict the function of protein and the information of protein interaction network is incomplete. In order to solve the problem, A Multi-Source Integration and Random Walk with Doubly Indexed Matrix (MSI-RWDIM) algorithm was proposed. The proposed algorithm used protein sequence, gene expression and protein-protein interaction for the prediction of protein function. The weighting networks were constructed from the data sources with their characteristics. A network, which was fused by the weighting networks, integrated with function correlation network to construct a doubly indexed matrix. Random walk was used to calculate annotation scores and predict protein function. The cross-validation experiments on Yeast show that MSI-RWDIM can achieve higher prediction accuracy, lower coverage and lower loss rate of function labels. The research results show that the overall performance of MSI-RWDIM is much better than commonly used k-nearest neighbor, transductive multi-label ensemble classifier and fast simultaneous weighting method.

Key words: multiple data integration, random walk, doubly indexed matrix, function correlation network, protein function prediction

摘要：

针对单一数据源预测蛋白质功能效果不佳以及蛋白质相互作用网络信息不完全等问题,提出一种多数据源融合和基于双重索引矩阵的随机游走的蛋白质功能预测(MSI-RWDIM)算法。该算法使用了蛋白质序列、基因表达和蛋白质相互作用数据预测蛋白质功能,并根据这些数据源特性构建相应的相互作用加权网络;然后融合各数据源加权网络并结合功能相关性网络构建双重索引矩阵,使用随机游走算法计算得分进而预测蛋白质功能。在酵母数据集的五折交叉验证中,MSI-RWDIM算法具有较高的准确率和较低的覆盖率,还可降低功能标签损失率。研究结果表明,MSI-RWDIM算法的总体性能优于常用的k-近邻、直推式多标签集成分类和快速同步加权方法。

关键词: 多数据源融合, 随机游走, 双重索引矩阵, 功能相关性网络, 蛋白质功能预测

CLC Number:

TP181

MENG Jun, ZHANG Xin. Protein function prediction based on doubly indexed matrix[J]. Journal of Computer Applications, 2015, 35(6): 1637-1642.

孟军, 张信. 基于双重索引矩阵的蛋白质功能预测[J]. 计算机应用, 2015, 35(6): 1637-1642.

References

[1] HAWKINS T, CHITALE M, LUBAN S, et al. PFP: Automated prediction of gene ontology functional annotations with confidence scores using protein sequence data [J]. Proteins: Structure, Function, and Bioinformatics, 2009, 74(3): 566-582.
[2] PUELMA T, GUTIERREZ R A, SOTO A. Discriminative local subspaces in gene expression data for effective gene function prediction [J]. Bioinformatics, 2012, 28(17): 2256-2264.
[3] MOOSAVI S, RAHGOZAR M, RAHIMI A. Protein function prediction using neighbor relativity in protein-protein interaction network [J]. Computational Biology and Chemistry, 2013, 43: 11-16.
[4] TROYANSKAYA O G, DOLINSKI K, OWEN A B, et al. A Bayesian framework for combining heterogeneous data sources for gene function prediction (in Saccharomyces cerevisiae) [J]. Proceedings of the National Academy of Sciences of the United States of America, 2003, 100(14): 8348-8353.
[5] BARUTCUOGLU Z, SCHAPIRE R E, TROYANSKAYA O G. Hierarchical multi-label prediction of gene function [J]. Bioinformatics, 2006, 22(7): 830-836.
[6] LAN L, DJURIC N, GUO Y, et al. MS-kNN: protein function prediction by integrating multiple data sources [J]. BMC Bioinformatics, 2013, 14(Suppl 3): S8.
[7] ZHANG X-F, DAI D-Q. A framework for incorporating functional interrelationships into protein function prediction algorithms [J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2012, 9(3): 740-753.
[8] JIANG J Q. Learning protein functions from bi-relational graph of proteins and function annotations [C]//WABI 2011: Proceedings of the 11th International Workshop, LNCS 6833. Berlin: Springer-Verlag, 2011: 128-138.
[9] HU P, JIANG H, EMILI A. Predicting protein functions by relaxation labelling protein interaction network [J]. BMC Bioinformatics, 2010, 11(Suppl 1): S64.
[10] CHOU K-C. Pseudo amino acid composition and its applications in bioinformatics, proteomics and system biology [J]. Current Proteomics, 2009, 6(4): 262-274.
[11] WANG J, 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.
[12] GUO X, GAO L, LIAO Q, et al. Long non-coding RNAs function annotation: a global prediction method based on bi-colored networks [J]. Nucleic Acids Research, 2013, 41(2): e35.
[13] ZHOU D, BOUSQUET O, LAL T N, et al. Learning with local and global consistency [EB/OL]. [2014-12-03]. http://131.107.65.14/en-us/um/people/denzho/papers/LLGC.pdf.
[14] CHATR-ARYAMONTRI A, BREIKREUTZ B J, HEINICKE S, et al. The BioGRID interaction database: 2013 update [J]. Nucleic Acids Research, 2013, 41(D1): D816-D823.
[15] DWIGHT S S, HARRIS M A, DOLINSKI K, et al. Saccharomyces Genome Database (SGD) provides secondary gene annotation using the Gene Ontology (GO) [J]. Nucleic Acids Research, 2002, 30(1): 69-72.
[16] MEWES H W, RUEPP A, THEIS F, et al. MIPS: curated databases and comprehensive secondary data resources in 2010 [J]. Nucleic Acids Research, 2010, 39(Suppl 1): D220-D224.
[17] REN Z, WANG L, FU Z, et al. Ensemble learning algorithm of muti-label classification based on Ranking Loss [J]. Journal of Computer Applications, 2013, 33(S1): 40-42. (任志博, 王莉莉, 付忠良, 等. 基于Ranking Loss的多标签分类集成学习算法 [J].计算机应用, 2013, 33(S1): 40-42.)
[18] YU G, RANGWALA H, DOMENICONI C, et al. Protein function prediction using multi-label ensemble classification [J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2013, 10(4): 1045-1057.
[19] MOSTAFAVI S, MORRIS Q. Fast integration of heterogeneous data sources for predicting gene function with limited annotation [J]. Bioinformatics, 2010, 26(14): 1759-1765.
[20] YOUNG B D, WEISS D I, ZURITA-LOPEZ C I, et al. Identification of methylated proteins in the Yeast small ribosomal subunit: A Role for SPOUT methyltransferases in protein arginine methylation [J]. Biochemistry, 2012, 51(25): 5091-5104.
[21] PROTCHENKO O, RODRIGUEZ-SUAREZ R, ANDROPHY R, et al. A screen for genes of heme uptake identifies the FLC family required for import of FAD into the endoplasmic reticulum [J]. Journal of Biological Chemistry, 2006, 281(30): 21445-21457.
[22] MARK K G, SIMONETTA M, MAIOLICA A, et al. Ubiquitin ligase trapping identifies an SCFSaf1 pathway targeting unprocessed vacuolar/lysosomal proteins [J]. Molecular Cell, 2014, 53(1): 148-161.
[23] KATO M, KITO K, OTA K, et al. Remodeling of the SCF complex-mediated ubiquitination system by compositional alteration of incorporated F-box proteins [J]. Proteomics, 2010, 10(1): 115-123.
[24] FOURATI Z, ROY B, MILLAN C, et al. A highly conserved region essential for NMD in the Upf2 N-terminal domain [J]. Journal of Molecular Biology, 2014, 426(22): 3689-3702.
[25] HLYNIALUK C, SCHIERHOLTZ R, VERNOOY A, et al. Nsf1/Ypl230w participates in transcriptional activation during non-fermentative growth and in response to salt stress in Saccharomyces cerevisiae [J]. Microbiology, 2008, 154(Pt 8): 2482-2491.
[26] SARAVANAN M, WUERGES J, BOSE D, et al. Interactions between the nucleosome histone core and Arp8 in the INO80 chromatin remodeling complex [J]. Proceedings of the National Academy of Sciences of the United States of America, 2012, 109(51): 20883-20888.
[27] MILLIGAN L, DECOURTY, L, SAVEANU C, et al. A yeast exosome cofactor, Mpp6, functions in RNA surveillance and in the degradation of noncoding RNA transcripts [J]. Molecular and Cellular Biology, 2008, 28(17): 5446-5457.
[28] ZHONG P P, MELCHER K. Identification and characterization of the activation domain of Ifh1, an activator of model TATA-less genes [J]. Biochemical and Biophysical Research Communications, 2010, 392(1): 77-82.