Table 2 Accuracy percentiles and characteristics of models from 50 cross-validation samples of DS1, m/z > 0
From: Performance of a genetic algorithm for mass spectrometry proteomics
Algorithm | Test Set Accuracy 25th, 75thPercentile | Median # Clusters | Median # Markers | Proportion of Perfect Chromosomes |
|---|---|---|---|---|
GA(P1 = 0, p2 = 0) | .96, .98 | 12 | 15.5 | 1.0 |
GA(0, .001) | .95, .99 | 12.5 | 11 | 1.0 |
GA(0, .002) | .95, .98 | 12.5 | 10 | 1.0 |
GA(.002, 0) | .96, .98 | 7 | 16 | .98 |
GA(.002, .001) | .97, .98 | 7 | 12.5 | 1.0 |
GA(.002, .002) | .96, .99 | 8.5 | 9 | .98 |
GA(.005, 0) | .96, .98 | 6 | 15 | .92 |
GA(.005, .001) | .96, .98 | 5 | 11 | .92 |
GA(.005, .002) | .96, .99 | 6 | 8 | .94 |
GA(.008, 0) | .96, .98 | 3 | 15 | .72 |
GA(.008, .001) | .97, .99 | 4 | 9 | .88 |
GA(.008, .002) | .97, .99 | 4 | 8 | .84 |
Best GA | .97, .99 | 6 | 12 | .94 |
Boosting | .99, 1.0 | NA | NA | NA |
PAM | .93, .97 | NA | NA | NA |