TABLE 2.
Summary of categorical agreement, very major errors, and major errors per compound class for WGS-based susceptibility prediction in comparison with broth microdilution susceptibility testinga
| Compound class | No. of compounds | No. of isolates | Categorical agreement (%) | VME (%) | ME (%) | TP | TN | FP | FN |
|---|---|---|---|---|---|---|---|---|---|
| Aminoglycosides | 3 | 462 | 924/965 (95.8) | 19/115 (16.5) | 22/850 (2.6) | 96 | 828 | 22 | 19 |
| Fluoroquinolones | 3 | 547 | 1067/1143 (93.4) | 34/382 (8.9) | 42/761 (5.5) | 348 | 719 | 42 | 34 |
| Tetracyclines | 1 | 128 | 116/128 (90.6) | 2/5 (40.0) | 10/123 (8.1) | 3 | 113 | 10 | 2 |
| Macrolides | 1 | 109 | 79/109 (72.5) | 26/69 (37.7) | 4/40 (10.0) | 43 | 36 | 4 | 26 |
| Folate pathway inhibitors | 2 | 431 | 486/559 (86.9) | 39/106 (36.8) | 34/453 (7.5) | 67 | 419 | 34 | 39 |
| Penicillins | 4 | 485 | 549/638 (86.1) | 38/217 (17.5) | 51/421 (12.1) | 179 | 370 | 51 | 38 |
| Cephalosporins (second generation) | 1 | 190 | 156/190 (82.1) | 9/68 (13.2) | 25/122 (20.5) | 59 | 97 | 25 | 9 |
| Cephalosporins (third generation) | 2 | 438 | 549/641 (85.6) | 44/187 (23.5) | 48/454 (10.6) | 143 | 406 | 48 | 44 |
| Cephalosporins (fourth generation) | 1 | 394 | 349/394 (88.6) | 18/104 (17.3) | 27/290 (9.3) | 86 | 263 | 27 | 18 |
| Carbapenems | 3 | 361 | 681/784 (86.9) | 34/139 (24.5) | 69/645 (10.7) | 105 | 576 | 69 | 34 |
a
Categorical agreement, very major errors (VME), and major errors (ME) were calculated from the overall number of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) obtained per compound class.