Skip to main content

Advertisement

Table 4 Specificity, sensitivity, phi correlation and misclassification rate for each model at detecting different α and (1-α) percentiles of extreme animals in the testing set within line C

From: Genome-wide prediction of discrete traits using bayesian regressions and machine learning

Parameter Method α (number of records)
   0.05 (7) 0.10 (24) 0.25 (80) 0.50 (104)
Specificity1 TBA 1 0.50 0.64 0.71
  BL 0 0.25 0.61 0.71
  RF 1 0.75 0.75 0.71
  L2B 1 1 0.96 0.98
  LhB 1 1 0.82 0.69
Sensitivity1 TBA 0.33 0.30 0.54 0.53
  BL 0.5 0.30 0.44 0.43
  RF 0.33 0.35 0.52 0.51
  L2B 0.17 0.20 0.15 0.15
  LhB 0.33 0.20 0.46 0.45
Phi correlation1 TBA 0.26 -0.16 0.17 0.24
  BL -0.35 -0.35 0.05 0.15
  RF 0.26 0.08 0.26 0.23
  L2B 0.17 0.20 0.17 0.24
  LhB 0.26 0.20 0.28 0.15
Misclassification rate (%)2 TBA 57 67 43 38
  BL 57 71 50 43
  RF 57 58 40 39
  L2B 71 67 56 44
  LhB 57 67 41 43
  1. 1Higher value is desirable; the best value for each percentile is in bold face;
  2. 2Lower value is desirable; the best value for each percentile is in bold face;
  3. TBA = Threshold Bayes A, BTL = Bayesian Threshold LASSO, RF = Random Forest; L2B = L2-boosting algorithm, LhB = Lh-boosting algorithm