Table 4. Results for Bioluminescent Repression of the Bacterium Genus Pseudomonas Dataset (Model Dataset 4) with Best Weighting (0.5–0–0.5) Combination as Obtained from the Retrospective Study.

model	division method	R2	Q2	MAE95%Train	QExt-F12	MAE95%Test	correct prediction in %	training set range	number of compounds outside ADa
1	sorted responseb	0.78	0.73	0.32	0.42	0.32	80.65	4.06	0
2		0.73	0.69	0.35	0.45	0.31	80.65		2 (45, 47)
3		0.80	0.74	0.28	0.64	0.30	87.10		1 (47)
4		0.78	0.73	0.29	0.55	0.30	83.87		1 (47)
5		0.80	0.74	0.29	0.66	0.31	83.87		1 (47)
6	Kennard–Stonec	0.67	0.61	0.32	0.62	0.40	81.25	4.06	0
7		0.72	0.60	0.32	0.66	0.39	84.38		0
8		0.68	0.65	0.32	0.62	0.39	78.13		0
9		0.67	0.60	0.33	0.62	0.40	81.25		0
10		0.69	0.63	0.32	0.63	0.39	78.13		0
11	modified-k-medoidsd	0.75	0.70	0.30	0.63	0.32	80.65	4.06	0
12		0.72	0.68	0.31	0.65	0.34	83.87		0
13		0.72	0.68	0.32	0.62	0.32	77.42		0
14		0.71	0.67	0.31	0.63	0.31	80.65		1 (64)
15		0.69	0.65	0.32	0.62	0.29	87.10		1 (64)

^aAD using standardization technique is used and compound ID mentioned under parenthesis.

^bN_Training = 73, N_Test = 31.

^cN_Training = 72, N_Test = 32.

^dN_Training = 73, N_Test = 31.