Skip to main content

Table 1 Training and test sets used in the four evaluation strategies

From: Automatic structure classification of small proteins using random forest

Strategy Training set Test set Training (Test) set size
1 DS1.69 DS1.73 Uniq 6929 (6606)
2 DS1.69 No - NA DS1.73 Unique - No - NA 4071 (4114)
3 DS1.69 No - NA DS1.69 NA 4071 (2858)
4 DS1.69 No - NA DS1.73 Unique - NA 4071 (4653)
  1. DS1.69, set of domain pairs from SCOP version 1.69, DS1.73 Uniq , set of domain pairs exclusive of SCOP version 1.73. DS1.69 No - NA and DS1.73 Unique - No - NA , are the respective DS1.69 and DS1.73 Uniq sets but without NA-pairs. DS1.69 NA and DS1.73 Unique - NA , are sets of only NA-pairs from DS1.69 and DS1.73 Uniq , respectively.