Table 4.
Evaluation of filter performance. This table presents the classification rates of the identified spectra for varying fractions of the significance threshold with and without retention time filtering. The model was trained using the vds3 dataset and the performance was measured on ds1 and ds2. In this context, tp stands for the number of true positives and fp for the number of false positives. The CR is tp divided by the sum of tp and fp.
Fraction of threshold | tp | fp | CR | tp with filter | fp with filter | CR with filter |
0.0 | 683 | 2572 | 0.2098 | 699 | 626 | 0.5275 |
0.1 | 682 | 2460 | 0.2171 | 692 | 602 | 0.5348 |
0.2 | 678 | 2260 | 0.2308 | 683 | 555 | 0.5517 |
0.3 | 669 | 1909 | 0.2595 | 668 | 483 | 0.5804 |
0.4 | 654 | 1410 | 0.3169 | 646 | 380 | 0.6296 |
0.5 | 624 | 868 | 0.4182 | 609 | 261 | 0.7000 |
0.6 | 575 | 474 | 0.5481 | 557 | 166 | 0.7704 |
0.7 | 516 | 235 | 0.6871 | 500 | 103 | 0.8292 |
0.8 | 468 | 125 | 0.7892 | 452 | 66 | 0.8726 |
0.9 | 420 | 72 | 0.8537 | 404 | 49 | 0.8918 |
1.0 | 366 | 46 | 0.8883 | 350 | 38 | 0.9021 |