化合物诱变活性的模糊数学方法量化计算及衡量  

Calculating and weighting the chemical mutigenicity activity by Fuzzy mathematics

作　　者：李永武[1] 叶志前[2] 陆金芳[2] 

机构地区：[1]浙江大学附属邵逸夫医院,浙江杭州310006 [2]浙江大学临床工程研究所,浙江杭州310006

出　　处：《计算机与应用化学》2007年第2期152-154,共3页Computers and Applied Chemistry

摘　　要：利用计算机将致癌化合物样本按KLN方法作结构结点编码、分解,获得机器可识别的化合物结构片段。根据模糊数学的隶属度概念和聚类方法,计算它对化合物诱变活性作用的隶属度,以衡量片段结构对化合物诱变活性的影响,在此量化基础上作模糊聚类,得特征的活性片段类,对化合物进行特征片段的匹配及匹配特征片段的量化平均,获得化合物活性的数值度量。计算表明:隶属度值的分布合理,片段的隶属度也确实反映了它们的活性情况,特征片段聚类后更确切的反映化合物的活性。同时,作为计算样本,当数量增加,效果更显著;由于片段分布的复杂性(错位、多样),用模糊性的度量可反映其本质。本文研究表明:模糊性的度量对于研究化合物构效关系是较好的方法。The carcinogenic chemicals sample are coded with structure＇ node by KLN methods on computer, and then broken up. The substructures of the chemicals the machine can recognize are gotten. By membership concept and cluster methods in the fuzzy mathematics, the membership of the substructure contributing to the chemical mutigenicity activity is calculated. It can weight the influence of the substructure of chemical on its activity, on which fuzzy cluster is made and end with characteristic substructure classes, then the sample are matched with the characteristic substructure by searching it＇s substructure, the result is averaged as one value to describe the chemical mutigenicity activity. All we do shows： the distribution of the membership value is reasonable, and the membership of substructure really reflects the activity of the chemical, characteristic substructure class can exactly show the activity of the chemical; as a computing sample,when the number of sample increases, the outcome is more effective. Because the distribution of substructure of the chemical is complex （diversity,reticular）, weighting the activity of the chemical by fuzzy mathematics is of its nature expression and more appropriate description. The study concludes that weighting activity with fuzzy mathematics is an effective way to study the QSAR （ quantitative substructure -activity relationships）.

关 键 词：模糊数学 诱变活性 片段 构效关系 隶属度 模糊聚类 

分 类 号：O631.11[理学—高分子化学] TQ051.9[理学—化学]