Abstract
<jats:p>Michaelis–Menten kinetic analysis is widely used to estimate enzymatic parameters, but the accuracy of these estimations depends not only on the regression method but also on the experimental design. Although nonlinear regression is generally considered the reference approach, linear transformations remain frequently used due to their simplicity. However, the influence of substrate distribution, noise, and experimental conditions on parameter estimation accuracy remains insufficiently characterized. In this study, four regression approaches (Lineweaver–Burk, Hanes–Woolf, Eadie–Hofstee, and nonlinear regression) were systematically evaluated using simulated Michaelis–Menten datasets generated under various experimental conditions. A total of 307 scenarios were investigated, corresponding to 10,350 simulated datasets covering nine kinetic parameter combinations. The impact of noise level, number of experimental values, repetition number, substrate distribution relative to Km, and inhibition models on Vmax, Km, and Ki estimations was assessed using relative error analysis. Linearization methods showed increased sensitivity to noise and substrate distribution, with Hanes–Woolf consistently exhibiting the poorest performance. Eadie–Hofstee and nonlinear regression provided the most reliable estimates across the tested conditions, although their performances were complementary depending on the parameter and inhibition model. Km estimation was generally more challenging than Vmax estimation, while increasing the number of experimental values or repetitions did not systematically improve accuracy when the substrate distribution was inadequate. Experimental validation confirmed that substrate distribution strongly influences parameter estimation, particularly when experimental values are mainly located above Km. Overall, this work demonstrates that robust enzymatic parameter estimation requires careful consideration of both regression strategy and experimental design. The results provide a framework for improving kinetic data acquisition and represent a step toward the development of more general guidelines for reliable parameter estimation beyond ideal Michaelis–Menten systems.</jats:p>