An ancova approach to normalize microarray data, and its performance to existing methods

Shih Huang Chan, Li Ju Chen, Nan Hwa Chow, Hiao Sheng Liu

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A microarray experiment includes many steps, and each one of them may include systematic variations. To have a sound analysis, the systematic bias must be identified and removed prior to the data being analyzed. Based on the M-A dependency observed by Dudoit et al. (2002), we suggest that, instead of using the lowess normalization, a new normalization method called ANCOVA be used for dealing with genes with replicates. Simulation studies have shown that the performance of the suggested ANCOVA method is superior to any of the available approaches with regards to the Fisher's Z score and concordance rate. We used a microarray data from bladder cancer to illustrate the application of our approach. The edge the ANCOVA method has over the existing normalization approaches is further confirmed through real-time PCR.

Original languageEnglish
Pages (from-to)257-268
Number of pages12
JournalJournal of Bioinformatics and Computational Biology
Volume3
Issue number2
DOIs
Publication statusPublished - 2005 Apr 1

All Science Journal Classification (ASJC) codes

  • Biochemistry
  • Molecular Biology
  • Computer Science Applications

Fingerprint Dive into the research topics of 'An ancova approach to normalize microarray data, and its performance to existing methods'. Together they form a unique fingerprint.

  • Cite this