模糊神经网络用于药物反相色谱结构与保留定量关系的研究

Application of Fuzzy Neural Networks in the Study of Quantitative Structure-Retention Relationship for RP-HPLC Analysis of Drugs

目的 应用模糊神经网络研究药物反相色谱的定量结构 -保留关系 (QSRR)。方法 在全面考察药物分子结构参数的基础上 ,使用模糊神经网络对 1 62个药物进行 QSRR的研究。结果 应用模糊神经网络成功地实现了溶剂强度和容量因子的预测 ,并获得了药物分子结构参数与反相色谱流动相溶剂强度和容量因子之间关系的模糊规则 ,实验验证的结果也比较理想。结论 模糊神经网络可较满意地用于药物反相色谱

AIM\ Fuzzy neural networks (FNNs) were applied to the study of q u antitative structure-retention relationship (QSRR) for RP-HPLC analysis of dru gs. METHODS\ Based on a thorough and systematic study on the molecula r structure of 162 drugs. FNNs were used to study QSRR of RP-HPLC. RESULT S\ Fuzzy rules reflecting the relationship among molecular structure, solven t strength and capacity factors were obtained automatically, and FNNs were succe ssfu lly used to predict the solvent strength and capacity factors (k′). The results of tests are satisfied. CONCLUSION\ The application of FNNs in the st udy of QSRR for RP-HPLC analysis had practical value in use.