Microarray 是一門新興的生物科技，可以同時監督數千基因的表現。在實驗過程中，有許多系統性的變異會影響所測得基因表現值，此等變異若未加以移除，將造成資料偏頗，使我們對基因解讀錯誤，而得到不正確的結論。如何控制microarray實驗變異，以確保microarray資料的品質，便成為分析microarray資料前必須要做的前置處理工作。由於Dudoit et al. (2002) 之lowess normalization並適用各種情況的資料，所以我們利用housekeeping genes的概念，提出一種normalization方法。並在文章中以模擬的方式評估所建議之normalization法與lowess之優劣，並以成功大學微生物免疫所提供的微陣列資料來說明所建議方法的應用。

　　Microarray is a new biotechnology which allows people to monitor the expression levels for thousands of genes simultaneously. There are many sources of systematic bias which may affect the measured gene expression levels during the fabrication of microarray. In order to making correct decision, this systematic bias has to be identified and be removed before analyzing data. In this paper, we pointed out the fault of lowess normalization and used the concept of housekeeping genes to propose a new normalization method. Simulation study shows that the performance of the suggested normalization method is superior to the available approaches. The proposed approach is illustrated with gene expression data provided by College of Medicine, NCKU.

[1] Dudoit, S., Fridlyand, J., and speed, T. P. (2002). Comparison of discrimination methods for the classification of tumors using gene expression data. Journal of the American Statistical Association, 97, 77-87.
[2] Dudoit, S., Yang, Y. H., Callow, M. J., Speed, T. P. and (2002). Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments. Statistica Sinica, 12, 111-139.
[3] Shin-Huang Chan, Li-Ju Chen, Nan-Hwa Chow and Hsiao-Sheng Liu (2005). An ancova approch to normalize microarray data, and its performance to existing methods. Journal of Bioinformatics and Computational Biology. Vol. 3 Issue 2, p257, 12p.
[4] D. L. Wilson, M. J. Buckley, C. A. Helliwell and l. W. Wilson (2002).New normalization methods for cDNA microarray data. Bioinformatics, Vol. 19 no. 11 2003, pages 1325-1332.
[5] Alexander Zien, Thomas Aigner, Ralf Zimmer and Thomas Lengauer (2001). Centralization : a new method for the normalization of gene expression data. Bioinformatics, Vol. 17 Suppl, 1 2001, pages s323-S331.
[6] 許惠婷(2004), M-A圖意涵及Normalization之商榷, 國立成功大學統計學研究所
[7] 陳麗如(2004), cDNA微陣列資料的新normalization方法, 國立成功大學統計學研究所