摘要
目的:考察改进的遗传算法(GA)在QSAR中的变量优选能力。方法:用自编程的改进的GA选择变量,PLS方法拟合模型,对7,9,10,11取代的喜树碱衍生物和取代苯环双三氮唑醇类抗真菌化合物两个数据组进行QSAR研究。结果:对两个数据组分别得到了一批较常规QSAR分析结果更优的方程。结论:GA作为一种启发式搜索算法以其高效快速、并行性的特点尤其适合于QSAR研究,但因其仅以统计指标为进化依据,故需对其实施必要的预处理及结果检查。
Abstract
OBJECTIVE:To investigate the ability of variables selection when an improved GA was used in QSAR study.METHODS:Two data sets:the camptothecin analogs with7,9,10 or 11 substituted and bistriazolols with substituted benzene ring was applied to an improved GA an alysis and PLS was used to fit models.RESULTS:Many better models were obtained than the result obtained by the standard techniques of variables selection in both data sets.CONCLUSION:GA,a heuristic searching algorithms,was suitable for QSAR because it is efficient,rapid and collateral.But it only depends on the statistic fingerposts,so some pretreatment and the checking to GA result is necessary.
关键词
遗传算法(GA) /
进化算法(EA) /
定量构效关系(QSAR) /
喜树碱 /
三唑醇类
{{custom_keyword}} /
Key words
genetic algorithms(GA) /
evolutionary algorithms(EA) /
QSAR /
camptothecin /
triazolols
{{custom_keyword}} /
朱杰;张万年;季海涛;周有骏;朱驹;吕加国.
一种改进的遗传算法及其在喜树碱和三唑醇类化合物的QSAR研究中的应用[J]. 中国药学杂志, 1999, 34(10): 694-697
Zhu Jie.
An improved genetic algorithms and its application to QSARs of camptothecinanalogs and bistriazolols[J]. Chinese Pharmaceutical Journal, 1999, 34(10): 694-697
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
1 HanschC，FujitaT．Ρ-σ-π Analysis．A method for the correlation of biological activity and chemical structure．J Am Chem Soc，1964，86(8)：1616．
2 Bυck T,Schwefel HP．An overview of evolutionary algorithms for parameter optimization．Euol Comput，1993，1(1)：1．
3 宋杰，张万年，周有骏，等．7，9，10，11位取代喜树碱衍生物的2D-QSAR研究．第二军医大学学报，1999，20(7)：449．
4 So SS，Karplus M．Evolutionary optimization in QSAR：an application of genetic neural networks．J Med Chem，1996，39：1521．
5 Rogers D，Hopfinger AJ．Application of genetic function approximation to quantitative structure-activity relationships and quantitative structure-property relationships．J Chem Inf Comput Sci，1994，34：854．
6 Luke BT．Evolutionary programming applied to the development of QSAR and QSPR．J Chem Inf Comput Sci，1994，34：1279．
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
“973”国家重点基础研究资助项目
{{custom_fund}}
