食品科学

• 生物工程 • 上一篇    下一篇

基于均匀设计与支持向量回归的拮抗辣椒链格孢菌解淀粉芽孢杆菌发酵条件优化

蔡文韬1,夏 菠1,*,夏延斌1,易有金1,任佐华2,袁泽民2,王立峰2   

  1. 1.湖南农业大学食品科技学院,湖南 长沙 410128;2.湖南农业大学生物安全科技学院,湖南 长沙 410128
  • 出版日期:2013-07-15 发布日期:2013-06-28
  • 通讯作者: 夏 菠
  • 基金资助:

    国家自然科学基金项目(31071738;31000827);湖南省教育厅重点项目(10A051);湖南省自然科学基金项目(09JJ-3032);
    湖南农业大学稳定人才科学基金项目(07WDl8)

Optimization of Fermentation Conditions for an Antagonistic Endophytic Bacillus amyloliquefaciens Strain L for Inhibition of Alternaria alternata in Chilli by Uniform Design and Support Vector Regression

CAI Wen-tao1,XIA Bo1,*, XIA Yan-bin1,YI You-jin1,REN Zuo-hua2,YUAN Ze-min2,WANG Li-feng2   

  1. 1. College of Food Science and Technology, Hunan Agricultural University, Changsha 410128, China;
    2. College of Biosaftey Science and Technology, Hunan Agricultural University, Changsha 410128, China
  • Online:2013-07-15 Published:2013-06-28
  • Contact: XIA Bo

摘要:

为了提高内生细菌解淀粉芽孢杆菌的抗菌代谢物质的产生,结合均匀设计(UD)与支持向量回归(SVR)发展 了一种新的配方优化方法UD-SVR,用来优化选择培养基成分和发酵罐条件。最优配方考虑9因素通过3轮混合水平 实验得出最佳优化条件为:淀粉0.5%、葡萄糖2%、酵母提取物0.4%、CaCl2 1.3%、KH2PO4 0.1%、30℃、36h、接 种量1%、pH7.0。发酵液稀释10倍抑制活性达到57.39%,比基准实验高出20.53%,UD-SVR明显优于参比模型,可 被广泛应用于实验设计和分析的处方优化过程中。

关键词: 优化, 均匀设计与支持向量机, 解淀粉芽孢杆菌, 抑菌活性

Abstract:

In order to increase the production of antimicrobial metabolites against Alternaria alternate by an antagonistic
endophytic strain L (Bacillus amyloliquefaciens), a novel experimental design and analysis approach, uniform design combined
with support vector regression (UD-SVR), was proposed to optimize culture medium components and fermenter conditions. Three
rounds of mixed level experiments of nine variables showed that the optimal culture medium consisted of starch 0.5%, glucose 2%,
yeast extract 0.4%, CaCl2 1.3% and KH2PO4 0.1% at pH 7.0, and the optimal fermentation conditions were 1% of inoculum amount,
30 ℃ and 36 h. The antimicrobial activity of a 10-fold dilution of the fermentation broth obtained under the optimized conditions was
57.39%, a 20.53% increase in comparison with the value from benchmark experiments. In conclusion, UD-SVR is more efficient than
the reference models and has more promising potential for wide applications in experimental design and optimization.

Key words: optimization, uniform design combined with support vector regression, Bacillus amyloliquefaciens, inhibitory activity

中图分类号: