Normalization of cDNA Microarray Data Using R-Language
Sang Cheol Kim1, In Uk Hwang2, In Young Kim¹, Sunho Lee³, Hyun Cheol Chung¹, Sun Young Rha¹ Byung Soo Kim²
1kimsc77@yonsei.ac.kr, Brain Korea 21 Project for Medical Science, Cancer Metastasis Research Center, Yonsei University College of Medicine; 2mzhwang@yonsei.ac.kr, Applied Statistics, Yonsei University
Normalization is an essential initial step for the statistical analysis of cDNA microarray data, which removes several sources of systemic variation irrelevant to the treatment effect. Yang et al. (2002, NAR) suggested several normalization methods including global normalization, intensity-dependent normalization, and within-print tip group normalization. However, these methods are technically difficult for biologists to perform the normalization of microarray data by themselves. We propose a user-friendly, R-based software program NOM-R, which implemented Yang et al.’s normalization procedures. In addition, this program can handle the repeated intensity values of replicated spots in a same array by taking their averages and normalize the two dye-swap experiments, simultaneously. We confirmed that the output data of NOM-R can be used effectively for further statistical analyses including SAM, clustering, PAM and R-MAT, a self-developed R-based microarray analysis tool.