A SOFTWARE FOR THE ESTIMATION OF BINDING PARAMETERS OF BIOCHEMICAL EQUILIBRIA BASED ON STATISTICAL PROBABILITY MODEL

An algorithm is proposed for the estimation of binding parameters for the interaction of biologically important macromolecules with smaller ones from electrometric titration data. The mathematical model is base d on the representation of equilibria in terms of probability concepts of statistical molecular thermodynamics. The refinement of equilibriu m concentrations of the components and estimation of binding parameter s (log site constant and cooperativity factor) is performed using sing ular value decomposition, a chemometric technique which overcomes the general obstacles due to near singularity. The present software is val idated with a number of biochemical systems of varying number of sites and cooperativity factors. The effect of random errors of realistic m agnitude in experimental data is studied using the simulated primary d ata for some typical systems. The safe area within which approximate b inding parameters ensure convergence has been reported for the non-sel f starling optimization algorithms. (C) 1998 Elsevier Science B.V.