Table 2.

Assessing the training set compound values for estimated and the experimental activities by Hypo1

Name	Fit value	Experimental IC50 (nmol/L)	Predicted IC 50 (nmol/L)	Errora	Experimental scale	Predicted scale
Molecule1	10.47	0.2	0.191	−1.04	+++	+++
Molecule2	8.58	5.2	14.9	2.86	+++	+++
Molecule3	8.40351	7.79	22.7	2.91	+++	+++
Molecule4	8.4847	7.85	18.83	2.39	+++	+++
Molecule5	7.6927	10	116.68	11.66	+++	+++
Molecule6	8.4082	10.3	22.46	2.18	+++	+++
Molecule7	8.95463	11.2	6.38	−1.75	+++	+++
Molecule8	8.44562	11.5	20.61	1.79	+++	+++
Molecule9	8.42254	23.9	21.73	−1.09	+++	+++
Molecule10	8.004	24	56.97	2.37	+++	+++
Molecule11	8.36497	24.7	24.816	1	+++	+++
Molecule12	8.50932	26	17.79	−1.46	+++	+++
Molecule13	8.20893	26.9	35.54	1.32	+++	+++
Molecule14	7.99567	29	58.08	2	+++	+++
Molecule15	8.48967	29.51	18.62	−1.58	+++	+++
Molecule16	8.36485	32.4	24.82	−1.3	+++	+++
Molecule17	8.19005	43.1	37.12	−1.16	+++	+++
Molecule18	7.79069	73	93.11	1.27	+++	+++
Molecule19	8.47375	101	19.31	−5.22	+++	+++
Molecule20	5.74562	15,400	10,329.50	−1.49	+	+
Molecule21	5.23885	16,600	33,178.10	1.99	+	+
Molecule22	5.71617	22,600	11,054.20	−2.04	+	+
Molecule23	5.7456	41,200	10,330.10	−3.98	+	+
Molecule24	5.74648	45,000	10,309.20	−4.36	+	+

^aError, ratio of the predicted activity (Pred IC50) to the experimental activity (Exp IC50) or its negative inverse if the ratio is < 1