Table 3.

Statistical analysis of the five largest clusters for InhA, FabI, FabV, and FabK.

Enzyme	Cluster	IntSim a	%Cmp b	nCl c	nCl/nCmp d	nCmp e	%Actives	nCmpE f	Q3 g	pIC50 h
FabI	1	0.30	46.7	22		208	77.4	445	6.6	9.4
FabI	2	0.43	13.3			59	72.9		6.9	8.1
FabI	3	0.41	9.0		0.049	40	47.5		5.8	6.9
FabI	4	0.11	7.2			32	75		6.7	6.7
FabI	5	0.29	4.7			21	33.3		5.9	6.6
FabK	1	0.30	88.2	3		45	53.3	51	7.6	8.6
FabK	2	0.08	9.8		0.059	5	20		5.1	5.3
FabV	1	0.40	100	1	0.067	15	33.3	15	6.4	7.0
InhA	1	0.28	16.4	77		162	38.9	990	6.2	7.2
InhA	2	0.52	11.5			114	36.0		7.0	9.7
InhA	3	0.48	7.4		0.078	73	45.2		6.5	7.7
InhA	4	0,20	6.9			68	95.6		8.4	8.7
InhA	5	0.47	6.6			65	10.8		5.1	6.0

^a IntSim—Intracluster similarity, ^b %Cmp—Proportion of compounds in an enzyme dataset, ^c nCl—Number of clusters, ^d nCl/nCmp—Number of clusters divided by the number of compounds, ^e nCmp—Number of compounds in a cluster, ^f nCmpE—Number of compounds in an enzyme dataset, ^g Q3—Third quartile of pIC₅₀ values in a cluster, ^h pIC₅₀—highest pIC₅₀ value in a dataset.