Skip to main content

Table 2 The number of correctly predicted template folds by random forest at the family level, superfamily level, and fold level under various ratios of negatives and positive training examples.

From: Improving protein fold recognition by random forest

Ratio Family Superfamily Fold
Top1 Top5 Top1 Top5 Top1 Top5
128:1 (Original) 469 508 275 344 131 187
100:1 457 505 271 335 119 183
75:1 466 503 273 330 118 181
50:1 459 504 276 329 120 174
25:1 452 504 270 324 114 181
1:1 190 287 81 157 49 107