Fig. 3: Performance evaluation with scarce data and unseen data. | Nature Communications

Fig. 3: Performance evaluation with scarce data and unseen data.

From: ColdstartCPI: Induced-fit theory-guided DTI predictive model with improved generalization performance

Fig. 3

a Performance comparison of ColdstartCPI with baselines under different scarce conditions on the BindingDB_AIBind dataset (number of data points n = 5). The error bar indicates the standard deviation. The missing rate indicates the percentage of non-training sets (validation and testing sets). The ratio of the validation set to the testing set is 1 to 9. b The spectral performance curves of AUC and AUPR for ColdstartCPI and baselines in the BindingDB_SPECTRA dataset. Source data are provided as a Source Data file.

Back to article page