Skip to main content
. 2014 Jul 1;2:e455. doi: 10.7717/peerj.455

Table 10. Extended performance measures for major cluster (class-I) predictions in each CDR-H1 and -H2 length (Phase 1).

Class-I predictions Statistics
CDR/Length True positives True negatives False positives False negatives Accuracy Precision Recall F-measure
DCP signatures
CDR-H1-13 173 6 21 1 0.89 0.89 0.99 0.94
CDR-H1-15 4 1 4 0 0.56 0.50 1.00 0.67
Canonical templates
CDR-H1-13 153 7 20 21 0.80 0.88 0.88 0.88
CDR-H1-15 3 2 3 1 0.56 0.50 0.75 0.60
DCP signatures
CDR-H2-9 41 0 0 0 1.00 1.00 1.00 1.00
CDR-H2-10 103 45 4 16 0.88 0.96 0.87 0.91
CDR-H2-12 7 0 1 0 0.88 0.88 1.00 0.93
Canonical templates
CDR-H2-9 27 0 0 14 0.66 1.00 0.66 0.79
CDR-H2-10 74 43 6 45 0.70 0.93 0.62 0.74
CDR-H2-12 7 0 1 0 0.88 0.88 1.00 0.93