As a rule, receptor-ligand assay data are fitted by logistic functions (4PL model, 5PL model, Feldman’s model). The preparation of the initial estimates for parameters of these functions is an important problem for processing receptor-ligand interaction data. This study represents a new mathematical approach to calculate the initial estimates more closely to the true values of parameters. The main idea of this approach is in using the modified linear least squares method for calculations of the parameters for the 4PL model and the Feldman’s model. In this study, the convergence of model parameters to true values is verified for the simulated data with different statistical scatter. Also, the results of processing real data for the 4PL model and the Feldman’s model are presented. A comparison is made of the parameter values calculated by the presented and a nonlinear method. The developed approach has demonstrated its efficiency in calculating the parameters of the complex Feldman”s models up to 4 ligands and 4 sites.

中文翻译：

---

以放射性配体测定数据为例，对配体结合测定数据进行计算机分析的方法

通常，受体-配体测定数据由逻辑函数拟合（4PL 模型、5PL 模型、Feldman 模型）。对这些函数的参数进行初始估计是处理受体-配体相互作用数据的一个重要问题。这项研究代表了一种新的数学方法，可以更接近于参数的真实值来计算初始估计值。这种方法的主要思想是使用改进的线性最小二乘法计算 4PL 模型和 Feldman 模型的参数。在本研究中，针对具有不同统计离散度的模拟数据，验证了模型参数与真值的收敛性。此外，还给出了 4PL 模型和 Feldman 模型处理真实数据的结果。对通过所提出的和非线性方法计算的参数值进行比较。所开发的方法在计算多达 4 个配体和 4 个位点的复杂 Feldman 模型的参数方面证明了其效率。