Free knot splines for biochemical data

The use of spline functions in the analysis of empirical two-dimensional (2-D) data (y(i), x(i)) is described. Spline functions are excellent empirical functions, which can be used with advantage instead of other ones, such as polynomials or exponentials. The knot location seen as variable value corresponds to classical parameter used to describe oxidation curves. An application on characterization of LDL oxidability shows free knot splines in a regression context.