Skip to main content

Table 3 Predictive accuracy of the models M3, (M3 + CHS) and (M3 + CHS − nHBint4)

From: Visual analytics in cheminformatics: user-supervised descriptor selection for QSAR methods

Model Predictive accuracy
M3 (Mn/MW, nHBint4, nHBint10, ETA_dEpsilon_B) R2 =  0.56
MAE =  4.03
RMSE = 6.22
M3 + CHS (Mn/MW, nHBint4, nHBint10, ETA_dEpsilon_B,CHS) R2 = 0.62
MAE = 3.43
RMSE = 5.89
M3 + CHS − nHBint4 (Mn/MW, nHBint10, ETA_dEpsilon_B,CHS) R2 = 0.69
MAE = 3.24
RMSE = 5.68
  1. The second column shows the predictive accuracy of the “best” model after applying 4-fold cross validation on three different methods (linear regression, decision trees, and neural networks). In this case, the best predictive accuracy for the three models was obtained by using a decision tree (M5P) and evaluating using 4-fold cross validation. The parameter setup and predictive accuracy for all methods is available in the Additional file 1: Table S3.