Table 1 Class prediction based evaluation of discrimination correctness expressed as percentage of the probe sets that were determined by one-way ANOVA with Benjamini and Hochberg false discovery rate (FDR) and of the probe sets corresponding to perfect match with E. coli K12 and Salmonella LT2 genomes.
From: Broad spectrum microarray for fingerprint-based bacterial species identification
Probe sets (FDR P-value) | Number of probes | Correct predictions | Percentage of correct predictions |
|---|---|---|---|
complete | 95,000 | 23/30 | 76.6% |
0.05 | 69,811 | 25/30 | 83.3% |
0.01 | 46,858 | 29/30 | 96.6% |
0.005 | 38,220 | 29/30 | 96.6% |
0.001 | 21,265 | 29/30 | 96.6% |
0.0005 | 15,439 | 29/30 | 96.6% |
0.0003 | 12,071 | 30/30 | 100% |
0.0001 | 6,379 | 30/30 | 100% |
0.00005 | 4,062 | 30/30 | 100% |
0.00001 | 1,160 | 30/30 | 100% |
0.000005 | 507 | 29/30 | 96.6% |
K12-match | 10,895 | 22/30 | 73.3% |
LT2-match | 9,849 | 23/30 | 76.6% |