Rank Difference Analysis of Microarrays (RDAM), a novel approach to statistical analysis of microarray expression profiling data

Table 3 Comparison of RDAM and SAM in the case of two, three or five replicates and 500 increased genes

		FDR50		FDR		True Positive		False Positive
R	p var	RDAM	SAM	RDAM	SAM	RDAM	SAM	RDAM	SAM
2R	0.3	55%	63%	33%	-	4	-	2	-
	0.1	25%	49%	21%	-	172	-	45	-
	0.05	9%	39%	11%	14%	281	23	35	4
	0.01	1%	26%	15%	23%	452	175	77	51
3R	0.3	42%	47%	17%	-	55	-	11	-
	0.1	7%	28%	16%	20%	344	156	64	40
	0.05	3%	19%	13%	18%	409	246	60	54
	0.01	0%	11%	14%	22%	473	381	80	109
5R	0.3	17%	21%	18%	25%	261	264	58	88
	0.1	2%	4%	15%	20%	455	395	81	101
	0.05	0%	2%	16%	20%	472	419	87	106
	0.01	0%	0%	19%	19%	474	443	111	105

Comparisons were made between two groups of two replicates. The first group is composed of synthetic data sets made from the experimental replicates Wt-t0a and Wt-t0b, in which 1000 genes (500 increased and 500 decreased) are changed, and the second group is composed of synthetic data sets without extra variation. Results are shown for increased genes. The first column (R), indicates the number of replicates used. The second column (p var), gives the p-value of the extra variation introduced to the changed genes. The third column displays the FDR50, and the fourth column (FDR) the real FDR corresponding to the set of genes selected at FDR = 20%. For the same level of selection, columns 5 and 6 give, respectively, the number of true and false positives.

ISSN: 1471-2105