Stability of gene selection methods for multiclass clssification

• Autorzy: Student S. , Fujarewicz K.
• Języki publikacji: EN

Abstrakty

EN

A big problem in applying DNA microarrays for classification is dimension of the dataset. Recently we proposed a gene selection method based on Partial Least Squares (PLS) for searching best genes for classification. The new idea is to use PLS not only as multiclass approach, but to construct more binary selections that use one versus rest and one versus one approaches. Ranked gene lists are highly instable in the sense, that a small change of the data set often leads to big change of the obtained ordered list. In this article, we take a look at the assessment of stability of our approaches. We compare the variability of the obtained ordered lists from proposed methods with well known Recursive Feature Elimination (RFE) method and classical t-test method. This paper focuses on effective identification of informative genes. As a result, a new strategy to find small subset of significant genes is designed. Our results on real cancer data show that our approach has very high accuracy rate for different combinations of classification methods giving in the same time very stable feature rankings.

Słowa kluczowe

EN

gene selection; partial least squares; stability selection; bootstrap .632+; multiclass classification; BBFR

PL

selekcja genów; metoda cząstkowych najmniejszych kwadratów; klasyfikacja wieloklasowa

• Wydawca: University of Silesia, Institute of Informatics, Computer Systems Department
• Czasopismo: Journal of Medical Informatics & Technologies
• Rocznik: 2010
• Tom: Vol. 15
• Strony: 101--107
• Opis fizyczny: Bibliogr. 15 poz., rys., tab.

Twórcy

autor

Student S.

• Silesian University of Technology, Automatic Control, Electronics and Computer Science; Automatic Institute, Akademicka 16, 44-100 Gliwice, Poland

autor

Fujarewicz K.

Bibliografia

• [1] BHATTACHARJEE A., Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses, 98(24): PNAS 2001, pp. 13790–13795.
• [2] BRAGA–NETO U, DOUGHERTY ER., Is cross–validation valid for small–sample microarray classification? 20(3), Bioinformatics 2004, pp. 374–380.
• [3] BOULESTEIX AL, SLAWSKI M., Stability and aggregation of ranked gene lists, Brief Bioinform (2009) 10: pp. 556–568.
• [4] DE JONG S., SIMPLS: An alternative approach to partial least squares regression, Vol. 18, Chemometrics Intell. Lab. Syst. 1993, pp. 25–263.
• [5] EFRON B., Bootstrap methods: another look look at the jackknife, Vol. 7, Annals of Statistics 1979, pp. 1–26.
• [6] EFRON B., TIBSHIRANI R., Improvements on cross–validation: the 632+ bootstrap method. Vol. 92, J. Amer. Statist. Assoc. 1997, pp. 548–560.
• [7] FERRARI F., BORTOLUZZI S., COPPE A., SIROTA A., SAFRAN M., Novel definition files for human GeneChips based on GeneAnnot. 8(446), BMC Bioinformatics 2007.
• [8] FUJAREWICZ K., A multigene approach to differentiate papillary thyroid carcinoma from benign lesions: gene selection using bootstrap–based Support Vector Machines. Vol. 14, Endocrine – Related Cancer 2007, pp. 809–826.
• [9] GELADI P., KOWALSKI BR., Partial Least–Squares Regresion: a Tutorial, Vol. 185, Analytica Chimica Acta 1986, pp. 1–17.
• [10] GUYON I., WESTON J., BARNHILL S., VAPNIK V., Gene selection for cancer classification using support vector machines, Vol. 46 , Machine Learning 2002, pp. 389–422.
• [11] HŐSKULDSSON A., PLS regression methods. Vol. 2(3), J. Chemometrics 1988, pp. 211–228.
• [12] MEINSHAUSEN N., BÜHLMANN P., Stability selection (with discussion). Journal of the Royal Statistical Society 2010: Series B, pp. 417−473.
• [13] NGUYEN DV., ROCKE DM., Tumor classification by partial least squares using microarray gene expression data. Vol. 18(1), Bioinformatics 2002, pp. 39–50.
• [14] STUDENT S., FUJAREWICZ K., Multiclass cancer classification and biomarker Discovery on microarray data. XV Krajowa Konferencja Zastosowań Matematyki w Biologii i Medycynie, Szczyrk 2009, pp. 130−136.
• [15] ZHANG T., LI C., OGIHARA M., A comparative study of feature selection and multiclass classification methods for tissue classification based on gene expression, Vol. 20(15), Bioinformatics 2004, pp. 2429–2437.