Skip to main content
. 2012 Apr 5;7(4):e34498. doi: 10.1371/journal.pone.0034498

Table 1. Statistical performance metrics for diagnostic primer sets.

Primer set: 393 901 781 396 237
TP: 21 21 21 21 21
FP: 3 4 2 2 6
TN: 30 29 31 31 27
FN: 0 0 0 0 0
Sensitivity (recall): 1.00 1.00 1.00 1.00 1.00
Specificity: 0.91 0.88 0.94 0.94 0.82
PPV (precision): 0.88 0.84 0.91 0.91 0.78
FPR: 0.09 0.12 0.06 0.06 0.18
FDR: 0.13 0.16 0.09 0.09 0.22
F-measure: 0.93 0.91 0.95 0.95 0.88

The confusion matrix counts (TP: true positive; FP: false positive; TN: true negative; FN: false negative) derived from the experimental validation tests against unseen isolates of E. coli are presented. No false negatives were identified by any individual primer set. Derived performance measures are also indicated (PPV: positive predictive value; FPR: false positive rate; FDR: false discovery rate; F-measure: 2×recall×precision/(recall+precision)). All primer sets amplify all positive examples, and have specificity between 82–94%, with 9–22% false discovery rate.