2D-quantitative structure–activity relationships model using PLS method for anti-malarial activities of anti-haemozoin compounds

Table 3 Evaluation results of 2D-QSAR models generated by PLS (Partial least squares) and PCR (Principal component regression) methods

	Model 1 (PLS)		Model 2 (PCR)
	With outliers	Without outliers	With outliers	Without outliers
Regression equation	pIC₅₀ = 0.04406 – 1.17564 × a_ICM + 4.80603 × GCUT_SLOGP_2 + 0.46880 × h_pavgQ + 0.00334 * PEOE_VSA-0 + 0.29021 × rsynth	pIC₅₀ = − 4.90988 + 1.98542 × a_ICM + 0.74756 × GCUT_SLOGP_2 + 0.59815 × h_pavgQ + 0.00837 × PEOE_VSA-0 – 0.12277 × rsynth	pIC₅₀ = 0.04406 – 1.17564 × a_ICM + 4.80603 × GCUT_SLOGP_2 + 0.46880 × h_pavgQ + 0.00334 × PEOE_VSA-0 + 0.29021 × rsynth	pIC₅₀ = −4 .95576 + 2.02008 × a_ICM + 0.19451 × GCUT_SLOGP_2 + 0.56938 × h_pavgQ + 0.00850 × PEOE_VSA-0 – 0.01478 × rsynth
R²	0.587642	0.745031	0.587642	0.738223
RMSE	0.392729	0.166261	0.392729	0.168465
Q²	0.025752	0.316410	0.025752	0.317759
R²_pred	0.773600	0.955400	0.773600	0.954200

ISSN: 1475-2875