Two designed experiments (latin square and reference sample) that used independent arrays to study the same tissue samples


Analysis of variance was performed on microarray data from two designed experiments that used independent arrays to study the same tissue samples.

In the first experiment (The Latin square experiment), an mRNA sample obtained from human liver tissue was compared to a second sample obtained from muscle tissue. The design used two arrays such that on array 1 the liver sample is assigned to the “red” dye and the muscle sample is assigned to the “green” dye. On array 2 the dye assignments were reversed.

In the second experiment (The reference sample experiment), an independent comparison was made of the same samples used in the first. Again two arrays were used, but in this case placenta was used as a “reference” sample. Each of the muscle and liver samples were directly compared to the placenta sample on one array such that the test samples (liver and muscle) were assigned to the green dye and the reference sample (placenta) was assigned to the red dye. The array and dye indices are assigned as before. See Kerr et al. (2000) for more details.




Kerr MK, Martin M, Churchill GA
J Comput Biol. 2000;7(6):819-37 PMCID: PMC

Saltwater marsh in Acadia National Park