Skip to main content

Table 5 External performance of single models for predicting single point logLD50 (mmol/kg)

From: SAR and QSAR modeling of a large collection of LD50 rat acute oral toxicity data

Model

R2

MAE

RMSE

#AD

%AD

rRF

0.590

0.432

0.585

1966

0.907

aiQSAR

0.651

0.390

0.541

1843

0.850

istKNN

0.628

0.387

0.545

1917

0.884

HPT-RF

0.620

0.398

0.541

1885

0.869

  1. For each model, the R2, the mean absolute error (MAE), the root-mean squared error (RMSE), the number (#AD) and the percentage (%AD) of predictions in AD are reported. The best values for each metric are italicized
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us