Table 2.

Assessment of gene classifiers

Data setPrediction rateSensitivitySpecificityP*OR
22-gene New Zealand classifier tested on German data
    NZ data (training; SVM)0.77 (0.66-0.86)0.53 (0.33-0.73)0.88 (0.77-0.95)0.0028.4 (3.5-21.4)
    NZ data minus four genes not found in German data were removed from NZ data set (training; SVM)0.720.380.870.011
    German data (test; SVM)0.71 (0.51-0.86)0.62 (0.32-0.86)0.79 (0.52-0.95)0.0025.9 (1.6-24.5)
19-gene German classifier tested on NZ data
    German data (training; 3-NN)0.84 (0.65-0.95)0.850.83<0.000124.1 (5.3-144.7)
    German data minus five genes not found in NZ data were removed from German data set (training; 3-NN)0.670.650.660.046
    NZ data (test; 3-NN)0.67 (0.55-0.78)0.42 (0.22-0.64)0.78 (0.65-0.89)0.0452.6 (1.2-6.0)
  • NOTE: Value in parentheses are 95% confidence interval.

  • Abbreviation: NZ, New Zealand.

  • * P values were calculated from 2,000 permutation of class labels.