基于压缩感知算法的基因表达谱数据分析  

Analysis of gene expression data based on compressive sensing algorithm

作　　者：任丛林[1] 王瑞平[1] 

机构地区：[1]北京交通大学计算机与信息技术学院,北京100044

出　　处：《北京生物医学工程》2013年第2期195-197,200,共4页Beijing Biomedical Engineering

摘　　要：目的基因表达谱数据分析是生物信息学领域最重要的研究内容之一。其可实现对不同病理分型的肿瘤的正确分类,对肿瘤诊断和治疗具有重大意义。方法本文应用压缩感知算法实现对胃癌基因表达谱数据的分类,运用训练数据构造冗余字典,采用随机分布的规范行矢量高斯矩阵构造感知矩阵,对训练数据和测试数据进行感知,利用正交l_2-范数算法对基因表达谱数据进行重建,在变换域中采用近邻法测试判断数据类别,与样本的实际类别相比较。结果实验结果表明,压缩感知算法与K均值聚类、SVM等其他分类算法相比有较高的分类正确率,且分类速度快,能避免特征选取的问题。结论本文方法对疾病的临床诊断和生物信息学研究有重要的参考和借鉴作用。Objective Analysis of gene expression data is one of the most important branches of bioinformatics research. Correctly classifying the samples with pathological classification is important for tumor diagnosis and treatment. Methods This paper introduces the compressive sensing algorism for the classification of gastric cancer gene expression data. The redundant dictionary is formed by using the training set,and the random matrix with Gaussian entries builds the sensing matrix with normal row vectors. In the test stage,the sensing matrix is projected onto the test vector, and the minimum 10-norm solution is computed with orthogonal l2-norm algorithm. The distance between the reconstruction vector and the train vector is employed to determine the class of the test data. Results Compared with classification methods of K-means, SVM and so on, the experimental results show compressive sensing algorism promising aspects as high accuracy and efficiency for gene expression data classification. Conclusions This method is important for clinical diagnosis and biomedical research.

关 键 词：压缩感知 稀疏化 冗余字典 基因表达谱 

分 类 号：R318.04[医药卫生—生物医学工程]