Evaluation of molecular descriptors for antitumor drugs with respect to noncovalent binding to DNA and antiproliferative activity

Table 5 Equations used to predict GI₅₀ values for DNA-binding drugs^a.

	Predictive equation	r	p	AdR²
All drugs^c	GI ₅₀ = 0.002(± 0.001)Complexity - 0.008(± 0.004)PSA + 5.688(± 0.591)	0.713	1.4 × 10^-2	0.426
Intercalators	GI ₅₀ = 0.742(± 0.219) *logKeq* + 0.002(± 0.000)Complexity + 0.008(± 0.002)PSA + 1.977(± 1.319)	0.894	1.6 × 10^-2	0.700
'M-region'	GI ₅₀ = 0.021(± 0.102) *logKeq* + 0.001(± 0.000)Complexity + 6.478(± 0.746)	0.984	1.8 × 10^-4	0.955

The predictive equations are presented for the three sets of drugs analyzed (All 15 drugs, Intercalators and 'M-region' compounds) described in the main text^b.
^a Obtained by multiple regression analysis. The predictive equations displayed are those statistically more significant for each set of predictors (actual p values, ANOVA test, are shown in the Table). Other details as in legend to Table 4.
^bThe cases (drugs) used in the calculations for each set were 15, 10 and 8 respectively.
^clogKeq was not a significant predictor for GI₅₀ for the set that contains the 15 (all) drugs.