Skip to main content

Table 5 GenSET performance was compare to six other machine learning programs using the average TPR (sensitivity), SPC (specificity), and AUC (area under the curve) values obtained with unfiltered attributes from the four organisms. GenSET 1 performed better than the other six programs in all areas whereas GenSET 2 gave better specificities

From: Computational approach to predict species-specific type III secretion system (T3SS) effectors using single and multiple genomes

Program TPR SPC AUC Reference
EffectiveT3 ~0.710 ~0.850 0.85–0.86 [10]
T3MM ~0.839 ~0.903 N/Ac [16]
SIEVE 0.9 0.88 0.95–0.96 [11]
BPBAac ~0.910 ~0.974 0.989 [12]
T3SEpre 0.927 0.945 N/A [13]
Meta-analytic ~0.90 ~0.90 0.993 [14]
GenSET 1a 0.938 b 0.979 0.993 This study
GenSET 2 0.617 0.984 0.968 This study
  1. aGenSET 1 and GenSET 2 denote average results from Phase 1 and Phase 2 respectively
  2. bBold number denotes the highest value in a given column
  3. c N/A Not applicable or not available