基于基因表达谱的前列腺癌分类方法研究

基于肿瘤基因表达数据,运用信息科学的方法和技术建立肿瘤的预测分类模型,对肿瘤的识别具有重要意义。针对该类问题,论文提出了一种利用支持向量机进行肿瘤分类与判别的方法。该方法在分析基因表达谱特征的基础上,首先对所有的基因进行聚类,并挑选出每一类的“代表基因”作为特征基因,然后采用支持向量机作为分类器进行肿瘤分类。论文以前列腺癌的基因表达谱数据为例进行分类实验并取得了良好的分类结果,表明了该方法的有效性和可行性。

Prediction of Cancer Based on Gene Expression Profiles

It is very significant for establishing tumor classification and prediction models using methodology and technique of information science with gene expression data,This paper proposes an approach for cancer molecular classification using support vector machines,The significantly correlated genes are selected using cluster and the signal to noise ratio.This paper uses a linear support vector machine as the classifier to classify samples and applies the method to prostate cancer dataset as a test case,The effectiveness and feasibility of the method proposed is proved by the results that all samples in the test set can be correctly classified.