基于XGBoost的RNA修饰位点的识别

Identifying the occurrence sites of different RNA modifications based on XGBoost

为了实现用一种方法更准确地识别几种不同类型的RNA修饰位点,提出了一种融合位置特异性单核苷酸及双核苷酸偏好特征的k-元组核苷酸组成(PseKNC)编码方式,并构建了一个基于XGBoost的RNA修饰位点的预测模型。通过交叉验证测试表明,该模型的识别准确率优于现有模型。

In order to identify several different types of RNA modification sites more accurately in one way,it proposes pseudo k-tuple nucleotide composition(PseKNC)encoding method that combines the position-specific mononucleotide and dinucleotide propensity characteristics.In addition,a new model for identifying RNA modification sites based on this encoding method is built.The cross-validation test shows that the recognition accuracy of this model is better than that of the existing model.