基于均匀设计与支持向量回归的发酵配方优化  

Experimental Design and Analysis of the Optimal Fermentation Medium Based on Uniform Design and Support Vector Regression

作　　者：袁哲明[1,2] 左斌[3] 谭泗桥[2] 谭显胜[2] 熊兴耀[1] 

机构地区：[1]湖南农业大学作物种质创新与资源利用国家重点实验室培育基地,湖南长沙410128 [2]湖南农业大学生物安全科学技术学院,湖南长沙410128 [3]湖南农业大学生物科学技术学院,湖南长沙410128

出　　处：《过程工程学报》2009年第1期148-152,共5页The Chinese Journal of Process Engineering

基　　金：国家自然科学基金资助项目(编号:30570351);教育部新世纪优秀人才计划基金资助项目(编号:NCET-06-0710);高等学校博士点基金资助项目(编号:200805370002)

摘　　要：结合均匀设计与支持向量回归,提出了一种新的配方优化实验设计与分析方法UD-SVR.将其应用于优化产谷氨酸脱羧酶大肠杆菌诱变株的培养基配方与发酵条件,在考虑9因素时仅通过2轮28个实验,酶活性(吸光度OD630)即由初始的1.528高效提升至2.303,明显优于二次多项式偏最小二乘回归等经验风险最小参比模型.UD-SVR为多因素多水平配方优化实验设计与分析提供了一套预测精度高、指导性强、可解释性好、优化高效的整体解决方案.Based on uniform design and support vector regression, UD-SVR, a novel experimental design and analysis method for the prescription optimization was proposed. It was applied to optimize the complicated fermentation medium including nine factors for a variant of Escherichia coli. The variant could produce glutamate decarboxylase which transformed gultamic acid into gamma aminobutyric acid in vitro. The optimization results of the medium by UD-SVR showed that OD630, an activity index of glutamate deearboxylase, increased from 1.528 in the initial median to 2.303 in the optimal medium after testing of 28 schemes. UD-SVR is more efficient than the reference models and has the potential to be widely used for experimental design and analysis of the prescription optimization.

关 键 词：均匀设计 支持向量回归 配方优化 

分 类 号：Q939.9[生物学—微生物学]