Table 2.

Pearson's Product Moment Correlation Coefficients among barley gene expression indices calculated from seven different methods.

Method	AD	MAS5.0	MBEI1	MBEI2	RMA	GCRMA	PDNN
AD	0.991 ± 0.005	0.975 ± 0.004	0.988 ± 0.001	0.985 ± 0.001	0.647 ± 0.007	0.791 ± 0.008	0.615 ± 0.007
MAS5.0	0.973	0.984 ± 0.009	0.961 ± 0.005	0.965 ± 0.003	0.619 ± 0.026	0.748 ± 0.024	0.583 ± 0.026
MBEI1	0.987	0.958	0.990 ± 0.005	0.988 ± 0.001	0.664 ± 0.011	0.797 ± 0.009	0.629 ± 0.008
MBEI2	0.985	0.963	0.988	0.990 ± 0.005	0.643 ± 0.006	0.774 ± 0.006	0.605 ± 0.006
RMA	0.647	0.616	0.662	0.643	0.993 ± 0.003	0.914 ± 0.002	0.939 ± 0.008
GCRMA	0.791	0.744	0.797	0.774	0.914	0.992 ± 0.005	0.923 ± 0.004
PDNN	0.614	0.581	0.628	0.604	0.940	0.923	0.992 ± 0.005

The upper triangle shows the mean and corresponding standard deviation of 24 correlation coefficients, r_ijk, (k = 1, 2,..., 24). r_ijk represents the correlation coefficient between 22,840 corresponding pairs of gene expression indices calculated by methods i and j from the k^th microarray sample. The diagonal cells show means and standard deviations of 24 correlation coefficients, r_imn (n = 1, 2, 3 and m = 1, 2,..., 8). For n = 1, 2, 3, r_imn corresponds to three correlation coefficients calculated from three possible pairs of replicates for the m^th cultivar (m = 1, 2,..., 8) using method i. The lower triangle shows the correlation coefficients between all pairs of 22,840 gene expression indices calculated from methods i and j across all k = 24 samples.

¹ MBEI PM only model

² MBEI PM-MM model