Prediction of Subcellular Localization of Eukaryotic Proteins Using Position-Specific Profiles and Neural Network with Weighted Inputs

Lingyun Zou; Zhengzhi Wang; Jiaomin Huang

doi:10.1016/S1673-8527(07)60123-4

Volume 34 Issue 12

Dec. 2007

Turn off MathJax

Article Contents

Article Navigation > Journal of Genetics and Genomics > 2007 > 34(12): 1080-1087

PDF( 298 KB)

Prediction of Subcellular Localization of Eukaryotic Proteins Using Position-Specific Profiles and Neural Network with Weighted Inputs

doi: 10.1016/S1673-8527(07)60123-4

College of Mechatronics and Automation, National University of Defense Technology, Changsha 410073, China

More Information

Corresponding author: E-mail address: zoulingyun@nudt.edu.cn (Lingyun Zou)
Received Date: 2007-03-12
Rev Recd Date: 2007-06-14

Available Online: 2007-12-31

Publish Date: 2007-12-20

Abstract

Abstract

Subcellular location is one of the key biological characteristics of proteins. Position-specific profiles (PSP) have been introduced as important characteristics of proteins in this article. In this study, to obtain position-specific profiles, the Position Specific Iterative-Basic Local Alignment Search Tool (PSI-BLAST) has been used to search for protein sequences in a database. Position-specific scoring matrices are extracted from the profiles as one class of characteristics. Four-part amino acid compositions and 1st–7th order dipeptide compositions have also been calculated as the other two classes of characteristics. Therefore, twelve characteristic vectors are extracted from each of the protein sequences. Next, the characteristic vectors are weighed by a simple weighing function and inputted into a BP neural network predictor named PSP-Weighted Neural Network (PSP-WNN). The Levenberg-Marquardt algorithm is employed to adjust the weight matrices and thresholds during the network training instead of the error back propagation algorithm. With a jackknife test on the RH2427 dataset, PSP-WNN has achieved a higher overall prediction accuracy of 88.4% rather than the prediction results by the general BP neural network, Markov model, and fuzzy k-nearest neighbors algorithm on this dataset. In addition, the prediction performance of PSP-WNN has been evaluated with a five-fold cross validation test on the PK7579 dataset and the prediction results have been consistently better than those of the previous method on the basis of several support vector machines, using compositions of both amino acids and amino acid pairs. These results indicate that PSP-WNN is a powerful tool for subcellular localization prediction. At the end of the article, influences on prediction accuracy using different weighting proportions among three characteristic vector categories have been discussed. An appropriate proportion is considered by increasing the prediction accuracy.
- subcellular localization,
- PSI-BLAST,
- position-specific scoring matrices,
- weighting function,
- BP neural network

FullText(HTML)

References(19)

References

[1]	Murphy, RF, Boland, et al. Towards a systematics for protein subcellular location: quantitative description of protein location patterns and automated analysis of fluorescence microscope images Proc Int Conf Intel Sys Mol Biol, 8 (2000),pp. 251-259
[2]	Reinhardt, A, Hubbard, et al. Using neural networks for prediction of the subcellular location of proteins Nucleic Acids Res, 26 (1998),pp. 2230-2236
[3]	Chou, KC, Elrod, et al. Protein subcellular location prediction Protein Eng, 12 (1999),pp. 107-118
[4]	Yuan, Z Prediction of protein subcellular locations using Markov chain models FEBS Lett, 451 (1999),pp. 23-26
[5]	Hua, S, Sun, et al. Support vector machine approach for protein subcellular localization prediction Bioinformatics, 17 (2001),pp. 721-728
[6]	Huang, Y, Li, et al. Prediction of protein subcellular locations using fuzzy kNN method Bioinformatics, 20 (2004),pp. 21-28
[7]	Emanuelsson, O, Nielson, et al. Predicting subcellular localization of proteins based on their N-terminal amino acid sequence J Mol Biol, 300 (2000),pp. 1005-1016
[8]	Nair, R, Rost, et al. Inferring subcellular localization through automated lexical analysis Bioinformatics, 18 (2002),pp. S78-S86
[9]	Lu, Z, Szafron, et al. Predicting subcellular localization of proteins using machine-learned classifiers Bioinformatics, 20 (2004),pp. 547-556
[10]	Altschul, SF, Madden, et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acids Res, 25 (1997),pp. 3389-3402
[11]	Park, KJ, Kanehisa, et al. Prediction of protein subcellular locations by support vector machines using compositions of amino acids and amino acid pairs Bioinformatics, 19 (2003),pp. 1656-1663
[12]	Jones, DT Protein secondary structure prediction based on position-specific scoring matrices J Mol Biol, 292 (1999),pp. 195-202
[13]	Ouali, M, King, et al. Cascaded multiple classifiers for secondary structure prediction Protein Sci, 9 (2000),pp. 1162-1176
[14]	Xie, D, Li, et al. LOCSVMPSI: a web server for subcellular localization of eukaryotic proteins using SVM and profile of PSI-BLAST Nucleic Acids Research, 33 (2005),pp. W105-W110
[15]	Holm, L Unification of protein families Curr Opin Struct Biol, 8 (1998),pp. 372-379
[16]	Oh, S, Lee, et al. Sensitivity analysis of single hidden layer neural networks with threshold functions Neural Networks, 6 (1995),pp. 1005-1007
[17]	Liu, YG, You, et al. A simple functional neural network for computing the largest and smallest eigenvalues and corresponding eigenvectors of a real symetric matrix Neurocomputing, 67 (2005),pp. 369-383
[18]	Zou, LY, Wang, et al. Prediction of non-coding RNA based on neural network with principal component analysis Journal of Biomedical Engineering Research, 26 (2007),p. 9
[19]	Yamashita, N, Fukushima, et al. On the rate of convergence of the Levenberg-Marquardt method Computing, 15 (2001),pp. 239-249