This paper presents an analysis of variance (ANOVA) approach for normalizing gene expression microarray data and estimating changes in gene expression. Spotted cDNA microarrays are used to measure the relative quantities of specific mRNAs in tissue samples. The authors demonstrate that ANOVA methods can be used to normalize microarray data and provide estimates of changes in gene expression that are corrected for potential confounding effects. This approach establishes a framework for the general analysis and interpretation of microarray data. The authors describe two experiments: a Latin square design and a reference sample design. In the Latin square design, mRNA samples from human liver and muscle tissues were compared using two arrays. In the reference sample design, the same samples were compared using placenta as a reference. The authors used ANOVA models to account for various sources of variation, including array, dye, and variety effects. They also used bootstrap methods to construct confidence intervals for the estimates of interest. The ANOVA models account for the complex interactions between arrays, dyes, and varieties. The authors found that the Latin square design had a particularly neat structure, with each of the sixteen possible effects completely confounded with one other effect. The reference sample design had more complex confounding structures, with some effects partially confounded. The authors used ANOVA to estimate the effects of gene expression differences between tissue samples. They found that the estimated gene effects and variety×gene interactions were significant. They also used bootstrap methods to construct confidence intervals for these estimates. The results showed that the estimated fold changes were significant at the 0.01 level. The authors concluded that ANOVA methods provide a robust and reliable approach for analyzing microarray data. They also noted that the results from the two experiments were reproducible, confirming the validity of the ANOVA approach. The authors emphasized the importance of accounting for potential confounding effects in microarray data analysis and the value of using ANOVA methods for this purpose.This paper presents an analysis of variance (ANOVA) approach for normalizing gene expression microarray data and estimating changes in gene expression. Spotted cDNA microarrays are used to measure the relative quantities of specific mRNAs in tissue samples. The authors demonstrate that ANOVA methods can be used to normalize microarray data and provide estimates of changes in gene expression that are corrected for potential confounding effects. This approach establishes a framework for the general analysis and interpretation of microarray data. The authors describe two experiments: a Latin square design and a reference sample design. In the Latin square design, mRNA samples from human liver and muscle tissues were compared using two arrays. In the reference sample design, the same samples were compared using placenta as a reference. The authors used ANOVA models to account for various sources of variation, including array, dye, and variety effects. They also used bootstrap methods to construct confidence intervals for the estimates of interest. The ANOVA models account for the complex interactions between arrays, dyes, and varieties. The authors found that the Latin square design had a particularly neat structure, with each of the sixteen possible effects completely confounded with one other effect. The reference sample design had more complex confounding structures, with some effects partially confounded. The authors used ANOVA to estimate the effects of gene expression differences between tissue samples. They found that the estimated gene effects and variety×gene interactions were significant. They also used bootstrap methods to construct confidence intervals for these estimates. The results showed that the estimated fold changes were significant at the 0.01 level. The authors concluded that ANOVA methods provide a robust and reliable approach for analyzing microarray data. They also noted that the results from the two experiments were reproducible, confirming the validity of the ANOVA approach. The authors emphasized the importance of accounting for potential confounding effects in microarray data analysis and the value of using ANOVA methods for this purpose.