Ding Jiarui, Shi Jinhong, Wu Fang-Xiang
Department of Mechanical Engineering, University of Saskatchewan, Saskatoon, S7N 5A9, SK Canada.
Int J Data Min Bioinform. 2011;5(1):73-88. doi: 10.1504/ijdmb.2011.038578.
In literature, hundreds of features have been proposed to assess the quality of tandem mass spectra. However, many of these features are irrelevant in describing the spectrum quality and they can degenerate the spectrum quality assessment performance. We propose a two-stage Recursive Feature Elimination based on Support Vector Machine (SVM-RFE) method to select the highly relevant features from those collected in literature. Classifiers are trained to verify the relevance of selected features. The results demonstrate that these selected features can better describe the quality of tandem mass spectra and hence improve the performance of tandem mass spectrum quality assessment.
在文献中，已经提出了数百种特征来评估串联质谱的质量。然而，这些特征中的许多在描述谱图质量时并不相关，并且它们会降低谱图质量评估的性能。我们提出了一种基于支持向量机的两阶段递归特征消除（SVM-RFE）方法，从文献中收集的特征中选择高度相关的特征。训练分类器以验证所选特征的相关性。结果表明，这些所选特征能够更好地描述串联质谱的质量，从而提高串联质谱质量评估的性能。
