Skip to main content

Table 1 Dataset distribution.

From: Predicting sumoylation sites using support vector machines based on various sequence features, conformational flexibility and disorder

  Training Set Test Set  
  Sites (%) Sites (%) TOTAL
Positive Sites (Consensus) 267 (3.23) 17 (3.18) 284
Positive Sites (Non-consensus) 90 (1.09) 7 (1.31) 97
Negative Sites (Consensus) 280 (3.39) 22 (4.11) 302
Negative Sites (Non-consensus) 7629 (92.29) 488 (91.39) 8117
Total 8266 (100) 534 (100) 8800
  1. A total of 381 experimentally verified sumoylation sites were divided into 4 categories. 357 of the positive sites formed the training set, in which 267 sites conformed to the consensus motif and 90 did not conform to the consensus motif. Remaining 24 positive sites formed the independent testing set, in which 17 conformed to the consensus motif and, 7 did not conformed to the consensus motif.