Summary of the seven real bacterial whole-genome sequencing data sets used in this study
| Type/data set ID | Species | No. of lineages | No. of isolates | No. of true SNVs | No. of false SNVs | Source |
|---|---|---|---|---|---|---|
| Training/validation | ||||||
| Zhao_2019 | B. fragilis | 11 | 556 | 227 | 492 | (Zhao et al. 2019) |
| Conwill_2022 | C. acnes | 50 | 854 | 1972 | 3840 | (Conwill et al. 2022) |
| Baker_2025 | C. acnes | 85 | 2007 | 3444 | 554 | (Baker et al. 2025) |
| S. epidermidis | 76 | 2015 | 1995 | 1420 | ||
| Test | ||||||
| Kim_2014 | S. aureus | 1 | 121 | 265 | — | (Kim et al. 2014) |
| Buddle_2024 | C. difficile | 1 | 96 | 55 | — | (Buddle et al. 2024) |
| Snitkin_2012 | K. pneumoniae | 1 | 20 | 31 | — | (Snitkin et al. 2012) |
| Giulieri_2022 | S. aureus | 1 | 16 | 11 | — | (Giulieri et al. 2022) |