The effectiveness of selected experiments improves with annotation specificity.
Difference between the average (1-AUC) of the 10-fold cross-validations, obtained for the selected set and for all experiments, for every term from the leaf node “GO:0009861: Jasmonic acid and ethylene dependent systemic resistance” up to the root. The greater the difference, the better the performance of the selected set.
Articles from PLoS ONE are provided here courtesy of
Public Library of Science