酸性和碱性酶模式识别的研究

Classification of acidic and alkaline enzymes via pattern recognition methods

了解酸性和碱性酶稳定性机理及在此基础上建立基于序列的模式识别方法对探讨其构效关系及酶的改造具有重要意义。本文采用主成分分析、偏最小二乘回归和BP神经网络3种方法对酸性和碱性酶进行模式识别。结果表明,基于主成分分析和偏最小二乘回归建立的线性方程能有效解释酸性和碱性酶稳定性机制,3种方法对训练集拟合的平均正确率分别为73.2%、87.0%和98.0%,建立了1种基于数学模型解释酶适应不同pH的分子机制及识别酸性和碱性酶的新方法。

To get the patterns of molecular adaptations to acidic and alkaline environment and its relationship with the physicochemical properties of amino acids is of great significance for understanding the actual structure-function relationship underlying the stability of enzymes. In this paper, pattern recognition of acidic and alkine proteins were studied through principle component analysis, partial least-square regression and BP neural network. The results showed that the fitting accuracy of the three methods was 73.2 %, 87.0 % and 98.0 %, respectively. A mathematical model was established and the biological meaning of it was expatiated on, a new method to discriminate the acidic and alkaline enzymes based on their sequences was established here.