摘要
主成分分析(PCA)采取降维思想,同时保持数据对方差贡献最大的特征,在畜牧生产上用于研究影响性状的变量,既简化变量个数,又获取足量信息,降低课题研究的复杂性。在全基因组关联分析(GWAS)中,PCA可用于校正群体分层,降低群体分层对关联结果的假阳性,通过PCA图可以看出研究群体是否有分层现象。本文主要对PCA的原理、分析软件以及在畜牧生产和GWAS中的应用加以综述。
Principal component analysis（PCA） takes the idea of dimensionality reduction and also maintains the characteristics of the largest contribution data to the difference. In livestock production, PCA is used to study variables of traits and expected to simplify the number of variables as well as obtain sufficient information to reduce the complexity of research. In genome-wide association analysis（GWAS）, PCA can be used to correct population stratification and reduce the false positive results of population stratification for association results. The PCA diagram can be shown whether the study population is stratified. In this paper, the principle of PCA, analysis software and its application in livestock production and GWAS are reviewed.
出处
《中国畜牧杂志》
CAS
北大核心
2017年第11期21-24,共4页
Chinese Journal of Animal Science
基金
河北省科技计划项目(15226301D)
关键词
主成分分析
群体分层
降维
假阳性
GWAS
Principal component analysis
Population stratification
Dimensionality reduction
False positive
GWAS
