Table 1. Validation of the Hidden Markov Models and Comparison to Current Pfam-B Models.
Target | Model | Hits | Correct hits | Missing hits |
CsgA/B (curli repeat) | Pfam (PF07012)1 | 354 | 354/354 (100%) | 47/401 (12%) |
CsgA/B (curli repeat) | This study | 379 | 374/379 (99%) | 27/401 (7%) |
CsgC | This study | 105 | 104/105 (99%) | 1/105 (1%) |
CsgD | This study | 16948 | 135/16948 (1%) | 1/136 (1%) |
CsgE | Pfam (PF10627) | 161 | 159/161 (99%) | 1/160 (1%) |
CsgE | This study | 159 | 159/159 (100%) | 1/160 (1%) |
CsgF | Pfam (PF10614) | 172 | 172/172 (100%) | 0/172 (0%) |
CsgF | This study | 172 | 172/172 (100%) | 0/172 (0%) |
CsgG | Pfam (PF03783) | 601 | 171/601 (26%) | 2/173 (1%) |
CsgG | This study | 266 | 171/266 (59%) | 2/173 (1%) |
CsgH | This study | 25 | 23/25 (92%) | 0/23 (0%) |
The size and location of the curli repeat region assigned by this model does not agree with the curli repeat regions described by [31].