论文部分内容阅读
支持向量机有许多优点:有效防止过拟和,适合大的特征空间,给定数据集的信息压缩。本文首次利用支持向量机从氨基酸组成来预测蛋白质的稳定性。总预测率可以达到80.64%,对嗜热蛋白质的预测率为82.50%,对嗜温蛋白质的预测率为80.29%从预测率可以验证氨基酸组成与蛋白质热稳定性成正相关的关系,支持向量机可以成为基于氨基酸组成预测蛋白质热稳定性的有效工具。
Support vector machines have many advantages: They can effectively prevent over-fitting, suitable for large feature space, and compress information in a given data set. In this paper, we first use the support vector machine to predict the stability of the protein from the amino acid composition. The total prediction rate can reach 80.64%, the prediction rate of thermophilic protein is 82.50%, the prediction rate of mesophilic protein is 80.29%. The predicted ratio can verify the positive correlation between amino acid composition and protein thermostability. Support Vector Machine Become an effective tool for predicting protein thermal stability based on amino acid composition.