Analytical Chemistry Department, Faculty of Pharmacy, Cairo University, Cairo, Egypt.
National Organization for Drug Control and Research, Giza, Egypt.
J AOAC Int. 2021 Jun 12;104(3):571-578. doi: 10.1093/jaoacint/qsaa176.
In many real-world situations there are many components in a mixture that produce an enormous amount of information.
The main task is to build up balanced models that convert these data into meaningful information to deal with. Hence, different chemometric models were applied for the analysis of data obtained from a mixture containing sofosbuvir, ledipasvir, velpatasvir, daclatasvir, and valacyclovir that were recently used internationally for their antiviral activity.
Partial Least Squares, Spectral Residual Augmented Classical Least Squares, and Concentration Residual Augmented Classical Least Squares designs were applied with and without variable selection procedure [Genetic Algorithm (GA)]. The methods were used for the quantitative analysis of the drugs in laboratory prepared mixtures and real market sample through handling the UV spectral data.
Robust models were obtained by applying GA. The proposed methods were found to be rapid, simple, and required no preliminary separation steps.
These models can be used on a routine basis in quality control laboratories or factories giving competitor results to those obtained by the reported methods.
The proposed models offer a powerful analytical alternative for laboratories that consider economic strategies in their requirements.
在许多实际情况中，混合物中有许多成分会产生大量信息。
主要任务是建立平衡模型，将这些数据转化为有意义的信息来处理。因此，不同的化学计量学模型被应用于分析含有索非布韦、来迪派韦索、维帕他韦、达卡他韦和缬更昔洛韦的混合物的数据，这些药物最近因其抗病毒活性而在国际上被使用。
偏最小二乘法、光谱残差增广经典最小二乘法和浓度残差增广经典最小二乘法设计，应用于和不应用变量选择程序[遗传算法（GA）]。这些方法用于通过处理紫外光谱数据，对实验室制备的混合物和实际市场样品中的药物进行定量分析。
通过应用 GA 得到了稳健的模型。研究发现，所提出的方法快速、简单，不需要预先分离步骤。
这些模型可以在质量控制实验室或工厂中常规使用，为那些考虑经济策略的实验室提供有竞争力的结果。
所提出的模型为那些在要求中考虑经济策略的实验室提供了一种强大的分析替代方案。
