期刊
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
卷 3, 期 1, 页码 51-57出版社
SPRINGER HEIDELBERG
DOI: 10.1007/s13042-011-0039-7
关键词
Autoregressive (AR) model; DNA microarray; Gene expression profiles; Singular spectrum analysis (SSA); Noise filtering
资金
- Hong Kong Grant Research Council [CityU 122607]
In genetic research, microarray chip carries thousands of genome expression profiles which allow biologists to analyze some of the developmental processes of life, such as biological reactions due to specific influences and so on. A main challenge of DNA microarray analysis is to separate the main gene expression from experimental noise. In order to ensure the accuracy of the following analysis, an effective noise filtering scheme is needed. In this paper, we propose a strategy to remove noise from gene expression profiles based on an autoregressive model based power spectrum analysis combined with singular spectrum analysis. This method helps us to determine the power spectrum effectively such that we can easily reconstruct the noise filtered time series signal.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据