State Key Laboratory of Non-food Biomass Enzyme Technology, National Engineering Research Center for Non-food Biorefinery, Guangxi Key Laboratory of Biorefinery, Guangxi Academy of Sciences, 98 Daling Road, Nanning, Guangxi, China.
Appl Biochem Biotechnol. 2013 Mar;169(6):1884-94. doi: 10.1007/s12010-013-0103-8. Epub 2013 Jan 24.
This is the continuation of our studies to use very basic information on enzyme to predict optimal reaction parameters in enzymatic reactions because the gap between available enzyme sequences and their available reaction parameters is widening. In this study, 23 features selected from 540 plus features of individual amino acid as well as a feature combined whole protein information were screened as independents in a 20-1 feedforward backpropagation neural network for predicting optimal pH in beta-glucosidase's hydrolytic reaction because this enzyme drew attention recently due to its role in biofuel industry. The results show that 11 features can be used as independents for the prediction, while the feature of amino acid distribution probability works better than the rest independents for the prediction. Our study paves a way to predict the optimal reaction parameters of enzymes based on the amino acid features of enzyme sequences.
这是我们利用酶的基本信息来预测酶促反应中最佳反应参数的研究的延续，因为可用酶序列与其可用反应参数之间的差距正在扩大。在这项研究中，从 540 多个单个氨基酸特征以及一个结合了整个蛋白质信息的特征中筛选出 23 个特征作为独立变量，用于预测β-葡萄糖苷酶水解反应的最佳 pH 值，因为由于其在生物燃料行业中的作用，这种酶最近引起了关注。结果表明，可以使用 11 个特征作为独立变量进行预测，而氨基酸分布概率特征比其他独立变量更适合预测。我们的研究为基于酶序列的氨基酸特征预测酶的最佳反应参数铺平了道路。
