Improving the prediction accuracy in classification using the combined data sets by ranks of gene expressions

BMC Bioinformatics

Table 5 Summarization of discretized data using ranks of gene expressons.

		Experimental groups by phenotypes
		P1	P2	P3	Marginal sum
Experimental group by rank (or ME^a) of gene expression	E1	n₁₁	n₁₂	n₁₃	r₁
	E2	n₂₁	n₂₂	n₂₃	r₂
	E3	n₃₁	n₃₂	n₃₃	r₃
Marginal sum		c₁	c₂	c₃	n

The significant genes can be selected by independency test between the phenotypes and gene expressions using this type of summarized data set. c_i,r_irepresent the marginal sums of the i^thcolumn and row, respectively. n represents total number of experiments. ^a: minimal entropy method

ISSN: 1471-2105