A Novel Method to Robust Tumor Classification Based on MACE Filter

  • Shulin Wang
  • Yihai Zhu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5755)

Abstract

Gene expression profiles consisting of thousands of genes can describe the characteristics of specific cancer subtype. By efficiently using the overall scheme of gene expression, accurate tumor diagnosis can be performed well in clinical medicine. However, faced many problems such as too much noise and the curse of dimensionality that the number of genes far exceeds the size of samples in tumor dataset, tumor classification by selecting a small set of gene subset from the thousands of genes becomes a challenging task. This paper proposed a novel high accuracy method, which utilized the global scheme of differentially expressed genes corresponding to each tumor subtype which is determined by tumor-related genes, to classify tumor samples by using Minimum Average Correlation Energy (MACE) filter method to computing the similarity degree between a test sample with unknown label in test set and the template constructed with training set. The experimental results obtained on two actual tumor datasets indicate that the proposed method is very effective and robust in classification performance.

Keywords

Gene expression profiles MACE filter tumor classification similarity degree DNA microarray 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Dabney, A.R.: Classification of microarrays to nearest centroids. Bioinformatics 21(22), 4148–4154 (2005)CrossRefGoogle Scholar
  2. 2.
    Wang, L.P., Chu, F., Xie, W.: Accurate cancer classification using expressions of very few genes. IEEE/ACM Transactions on computational biology and bioinformatics 4(1), 40–53 (2007)CrossRefMathSciNetGoogle Scholar
  3. 3.
    Huang, H.L., Lee, C.C., Ho, S.Y.: Selecting a minimal number of relevant genes from microarray data to design accurate tissue classifiers. BioSystems 90(1), 78–86 (2007)CrossRefGoogle Scholar
  4. 4.
    Sreekumar, J., Jose, K.K.: Statistical tests for identification of differentially expressed genes in cDNA microarray experiments. Indian Journal of Biotechnology 7(4), 423–436 (2008)Google Scholar
  5. 5.
    Deng, L., Ma, J.W., Pei, J.: Rank sum method for related gene selection and its application to tumor diagnosis. Chinese Science Bulletin 49(15), 1652–1657 (2004)MATHMathSciNetGoogle Scholar
  6. 6.
    Li, L.P., Darden, T.A., Weinberg, C.R., Levine, A.J., Pedersen, L.G.: Gene assessment and sample classification for gene expression data using a genetic algorithm/k-nearest neighbor method. Combinatorial Chemistry & High Throughput Screening 4(8), 727–739 (2001)Google Scholar
  7. 7.
    Zhou, X., Tuck, D.P.: MSVM-RFE: extensions of SVM-RFE for multiclass gene selection on DNA microarray data. Bioinformatics 23(9), 1106–1114 (2007)CrossRefGoogle Scholar
  8. 8.
    Troyanskaya, O.G., Garber, M.E., Brown, P.O., Botstein, D., Altman, R.B.: Nonparametric methods for identifying differentially expressed genes in microarray data. Bioinformatics 18(11), 1454–1461 (2002)CrossRefGoogle Scholar
  9. 9.
    Lehmann, E.L.: Non-parametrics: Statistical methods based on ranks, Holden-Day, San Francisco (1975)Google Scholar
  10. 10.
    Wilcoxon, F.: Individual comparisons by ranking methods. Biometrics 1, 80–83 (1945)CrossRefGoogle Scholar
  11. 11.
    Kruskal, W.H., Wallis, W.A.: Use of ranks in one-criterion variance analysis. Journal of the American Statistical Association 47(260), 583–621 (1952)MATHCrossRefGoogle Scholar
  12. 12.
    Mahalanobis, A., Kumar, B.V.K., Casasent, D.: Minimum average correlation energy filters. Appl. Opt. 26, 3633–3640 (1987)CrossRefGoogle Scholar
  13. 13.
    Kumar, B.V.: Tutorial survey of composite filter designs for optical correlators. Appl. Opt. 31, 4773–4801 (1992)CrossRefGoogle Scholar
  14. 14.
    Kumar, B.V., Savvides, V.M.K., Xie, C., Thornton, J., Mahalanobis, A.: Biometric verification using advanced correlation filters. Appl. Opt. 43, 391–402 (1992)CrossRefGoogle Scholar
  15. 15.
    Kumar, B.V.: Minimum variance synthetic discriminant functions. Opt. Soc. Am. A 3, 1579–1584 (1986)CrossRefGoogle Scholar
  16. 16.
    Yeoh, E.J., Ross, M.E., Shurtleff, S.A., Williams, W.K., Patel, D., Mahfouz, R., Behm, F.G., Raimondi, S.C., Relling, M.V., Patel, A., Cheng, C., Campana, D., Wilkins, D., Zhou, X., Li, J., Liu, H., Pui, C.H., Evans, W.E., Naeve, C., Wong, L., Downing, J.R.: Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell 1(2), 133–143 (2002)CrossRefGoogle Scholar
  17. 17.
    Khan, J., Wei, J.S., Ringner, M., Saal, L.H., Ladanyi, M., Westermann, F., Berthold, F., Schwab, M., Antonescu, C.R., Peterson, C., Meltzer, P.S.: Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nature Medicine 7(6), 673–679 (2001)CrossRefGoogle Scholar
  18. 18.
    Deutsch, J.M.: Evolutionary algorithms for finding optimal gene sets in microarray prediction. Bioinformatics 19(1), 45–52 (2003)CrossRefGoogle Scholar
  19. 19.
    Tibshirani, R., Hastie, T., Narasimhan, B., Chu, G.: Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proceedings of the National Academy of Sciences of the United States of America 99(10), 6567–6572 (2002)CrossRefGoogle Scholar
  20. 20.
    Dabney, A.R., Storey, J.D.: Optimality driven nearest centroid classification from genomic data. PLoS ONE 2(10), e1002 (2007), doi:10.1371/journal.pone.0001002CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Shulin Wang
    • 1
    • 2
  • Yihai Zhu
    • 1
    • 3
  1. 1.Intelligent Computing Lab, Hefei Institute of Intelligent MachinesChinese Academy of ScienceHeifeiChina
  2. 2.School of Computer and CommunicationHunan UniversityChangshaChina
  3. 3.Department of AutomationUniversity of Science and Technology of ChinaHefeiChina

Personalised recommendations