A Derived QSAR Model for Predicting Some Compounds as Potent Antagonist against Mycobacterium tuberculosis: A Theoretical Approach

Table 6

Validation parameters for each model using multilinear regression (MLR).

S/NO	Validation Parameters	Formula	Threshold	Model 1	Model 2	Model 3	Model 4
Internal Validation
1	Friedman LOF			0.03167	0.03253	0.03561	0.04567
2	R-squared			0.9265	0.8765	0.8454	0.8123
3	Adjusted; R-squared			0.9045	0.8464	0.8277	0.7800
4	Cross validated R-squared (			0.8512	0.8154	0.7574	0.7245
5	Significant Regression			Yes	Yes	Yes	Yes
6	Critical SOR F-value (95%)			3.6465	3.6542	3.75443	3.8743
7	Replicate points			0	0	0	0
8	Computed observed error			0	0	0	0
9	Min expt. error for non-significant LOF (95%)			0.03432	0.0354	0.04632	0.0485
Model Randomization
10	Average of the correlation coefficient for randomized data ()			0.3866	0.3265	0.4644	0.4875
11	Average of determination coefficient for randomized data (			0.1465	0.1843	0.2541	0.2533
12	Average of leave one out cross-validated determination coefficient for randomized data ( )			-1.3325	-1.3522	-1.4023	-1.4854
13	Coefficient for Y-randomization (c			0.7443	0.7103	0.6587	0.5873
External validation
14	Slope of the plot of Observed activity against Calculated activity values at zero intercept		0.85<k<1.15	1.0016	1.04732	1.0054	1.1134
15	Slope of the plot of Calculated against Observed activity at zero intercept		0.85<k<1.15	0.81233	0.9432	0.6432	0.96433
16			<0.3	0.01643	0.07433	0.05322	0.04324
17			<0.1	0.00243	0.00573	0.07843	0.0643
18			<0.1	0.05332	0.06453	0.07637	0.8633
19				0.8034	0.75433	0.6765	0.6123