Randomization test is employed as a gene selection method.
The method can evaluate the significance of a gene by a statistic
of the regression coefficients in a series of random PLSDA models.
Therefore, a few of the significant genes can be selected from the
thousands or more genes in an expression data. With repetition
of the calculations, the frequency number of a gene can be further
used as a criterion to evaluate its significance. Four datasets of
prostate cancer dataset, lung cancer dataset, leukemia dataset
and NSCLC dataset are investigated by the method. 18, 4, 9 and 7
significant genes are identified, respectively, and the rationality
of the results is validated by MLR modeling and PCA. Compared
with the results obtained in previous studies, the superiority of
the method is proved. Therefore, the method may be an alternative
tool for classification using the expression data.
Randomization test is employed as a gene selection method.The method can evaluate the significance of a gene by a statisticof the regression coefficients in a series of random PLSDA models.Therefore, a few of the significant genes can be selected from thethousands or more genes in an expression data. With repetitionof the calculations, the frequency number of a gene can be furtherused as a criterion to evaluate its significance. Four datasets ofprostate cancer dataset, lung cancer dataset, leukemia datasetand NSCLC dataset are investigated by the method. 18, 4, 9 and 7significant genes are identified, respectively, and the rationalityof the results is validated by MLR modeling and PCA. Comparedwith the results obtained in previous studies, the superiority ofthe method is proved. Therefore, the method may be an alternativetool for classification using the expression data.
การแปล กรุณารอสักครู่..
