遗传算法优化尿样核苷毛细管电泳分析条件

An approach to optimizing the separation of capillary electrophoresis with genetic algorithm

目的:研究建立一种基于遗传算法的尿样核苷类成分的毛细管电泳条件优化方法。方法:用中心组成试验设计系统考察硼砂、SDS、pH 和电压等因素对尿样核苷类成分电泳分离结果的影响。采用色谱指数方程对分离结果进行评价,并将其作为适应度函数。运用实数编码的遗传算法进行全局寻优,获取最优分析条件。结果:在优化的分析条件(15.7 mmol·L^(-1)硼砂,250.0 mmol·L^(-1)SDS,pH 9.60,电压14.8 kV)下,各核苷类成分在12 min 内得到较好分离。结论:本文方法准确可靠,适用于毛细管电泳分析条件优化。

Objective：A novel method based on genetic algorithm（GA） was proposed to optimize the separation of nucleosides by capillary electrophoresis （CE）. Method：Central composite design was implemented to investigate the influence of the factors（ borate, SDS, pH and applied voltage）on the separation performance. Chromatographic exponential function（CEF） was used to evaluate the quality of the CE separation and worked as the fitness function in optimization. Float - coded GA was utilized for the solution of the optimization problem. Result： The optimized result （borate 15.7 mmol · L^-1 ,SDS 250. 0 mmol · L^-1 ,pH 9.60,vohage 14. 8 kV） obtained from GA provided the best separation with regard to the resolution and analysis time. Conclusion：GA was an effective tool for the optimization of the separation of CE.