Evaluating methods for ranking differentially expressed genes applied to microArray quality control data

BMC Bioinformatics

Table 1 Average AUC values for AFX data

Technology	TAQ	TAQ	TAQ	TAQ	GEX	GEX	GEX	GEX
Comparison	AB	AB	CD	CD	AB	AB	CD	CD
Metric	FDR	AD	FDR	AD	FDR	AD	FDR	AD
#DEGs	817	553	767	351	144	106	114	60
WAD	69.91	75.74	71.41	77.69	75.74	78.75	76.54	80.69
AD	67.73	77.45	69.19	81.24	70.77	76.60	71.26	78.96
FC	67.73	77.44	69.17	81.20	70.79	76.61	71.35	78.95
RP	67.37	77.43	68.85	81.78	70.61	76.36	71.19	79.46
modT	68.74	76.98	70.19	79.95	73.12	77.67	73.53	80.77
samT	67.96	77.52	69.49	81.05	71.26	76.87	71.93	79.61
shrT	68.64	77.11	70.16	80.08	72.90	77.64	73.39	80.68
ibmT	69.26	76.85	70.70	79.57	74.27	78.26	74.87	81.33

Average AUC values across six test sites. Highest AUC values for each case are in bold. A total of 8 cases (2 assay technologies × 2 comparisons × 2 metrics) are shown. For example, AUC value for samT (77.52) was the highest in TAQ_AB_AD case, where 553 genes were defined as DEGs.

ISSN: 1471-2105