Relationship between Metabolic Fluxes and Sequence-Derived Properties of Enzymes

Table 3

Elements and the statistical indices for multiple linear regression models which link the values of metabolic fluxes and the average AA properties of the yeast Saccharomyces cerevisiae enzyme sequences expressed according to the AAindex scales.

Model	Dependent^a; variable	Parameters^b	Regression coefficient; (standardized value)	S.E.	value	value	, %; , %	VIF^c
III	Metabolic flux (Teusink's model)	Constant	1206.180	188.077	6.41	0.0000	94.02
							93.19
		()^2	−61514.800 (−0.865)	3379.980	−18.20	0.0000		1.36
			679.041 (0.400)	78.424	8.66	0.0000		1.29
		()^2	−17.034 (−0.621)	1.696	−10.05	0.0000		2.30
		()^2	−0.070 (−0.378)	0.014	−5.12	0.0000		3.28
			−4.820 (−0.336)	0.847	−5.69	0.0000		2.10
IV	Metabolic flux (Hynne's model)	Constant	2450.460	187.878	13.04	0.0000	90.25
							88.41
		()^2	−0.181 (−1.570)	0.013	−13.88	0.0000		4.86
			−2592.690 (−1.202)	182.567	−14.20	0.0000		2.72
			−466.789 (−0.358)	83.438	−5.60	0.0000		1.55
		()^2	−10.664 (−0.610)	1.388	−7.68	0.0000		2.39
		()^2	−17453.200 (−0.423)	2559.790	−6.82	0.0000		1.46
			276.424 (0.360)	49.933	5.54	0.0000		1.61
			2.648 (0.349)	0.594	4.45	0.0001		2.33

Represent the mean values of metabolic fluxes within the range of external glucose concentrations as specified in the “Material and Methods.”
^bElements of multiple linear regression which represent the average AA property, as specified in Table 1, of the yeast Saccharomyces  cerevisiae enzyme sequences and the constant (intercept) of equation.
^cThe variance inflation factor which indicates the impact of collinearity between the independent variables [22].