基于支持向量聚类的肿瘤表达谱分型识别算法

对肿瘤样本进行准确的分型识别是有效治疗肿瘤的前提。首先,利用方差滤波方法选择肿瘤表达谱中具有最大方差的部分基因作为识别特征集,然后,利用支持向量聚类对肿瘤表达谱进行分型识别。针对多类型样本情况和支持向量聚类中出现的孤立点聚类问题,分别提出了有效的解决办法。对两个肿瘤表达谱数据的测试结果显示,基于支持向量聚类的方法能够准确地对肿瘤样本进行分型识别,同时能够自动发现肿瘤样本真实的亚型数量。

Class Discovery of Tumor Microarray Expression Profiling Using Support Vector Clustering

It's very important for effectively curing tumor to discover correct classes of tumor.Firstly,a few of genes with the largest variance as characters for class discovery of tumor were selected by variance filtering.And then the support vector clustering was used to discover classes of tumor.During this course,the problems of multi-classes samples clustering and isolated points clustering were respectively solved by effective methods.The results on two tumor microarray datasets indicate that the support vector clustering can accurately discover classes of tumor samples and automatically find true class number of tumor samples.