cDNA微陣列資料中基因的變異數通常會不一致，而與均數呈現某種函數關係，此種現象造成選取出來的顯著基因通常有高的偽陽率。為處理此問題，Durbin et al.（2002）和Inoue et al.（2004）建立基因表現值的模型，藉由此模型推導出變異數和均數間的函數關係，之後透過此函數關係導出變異數穩定轉換函數，達成變異數的穩定化。

　　上述文獻所探討的基因表現值都是單一顏色光的強度，本文所探討的基因表現值形式為log(R/G)，即兩個顏色之表現值差異。本研究將不透過基因表現值的模型導出變異數與均數的函數關係，而是從實際資料的變異數對均數的散佈圖中找出兩者的關係，而後以無母數變異數穩定轉換的方式從事變數轉換。結果發現無論在統計模擬或是實例應用中，在挑選顯著基因上所提的方法都有不錯的表現。

For cDNA microarray data, the variance of gene is usually not the same and depends on its mean. Durbin et al. (2002) and Inoue et al. (2004) established the one-color gene expression model and obtain the relationship between variance of gene and its mean. They then derive the variance-stabilizing transformation function to stabilize the variance of the genes.

　In this article, we consider the two-color design and use nonparametric regression approach to stabilize the variance of gene expression level. We first, by applying lowess method, find the relationship between variance and mean of gene expression from scatter plot of variance versus mean, then use exponential function to approximate the relationship between variance and mean in a small region. Simulation study and real data analysis show that the performance of the suggested method is comparable to the parametric variance stabilization approach when the variance function is known.

Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rates: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society, Ser. B 57, 289-300.

Dudoit, S, Yang, Y. H., Callow, M. J. and Speed, T. P. (2002). Statistical methods for identifying differentially express genes in replicate cDNA microarray experiment. Statistica Sinica, 12, 111-139.

Durbin, B. P., Hardin, J. S. Hawkins, D. M. and Rocke, D. M. (2002). A variance-stabilizing transformation for gene-expression microarray data. Bioinformatics, 18, 105-110.

Huber, W., Heydebreck, A. V., Sultmann, H., Poustka, A. and Vingron, M. (2002). Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics, 18, 96-104.

Inoue, M., Nishimura, S. I., Hori, G., Nakahara, H., Saito, M., Yoshihara, Y. and Amar, S. I. (2004). Improved parameter estimation for variance-stabilizing transformation of gene-expression microarray data. Journal of Bioinformatics and Computational Biology, 2, 669-679.