摘要
基于性状—标记多元回归分析,提出在各条染色体上进行的QTL定位方法,采用“伪似然比”统计量进行显著性测验,并应用多元逐步回归对所检测到的QTL进行进一步地筛选,这不仅简化计算,而且明显提高统计功效。同时给出模拟例子,说明该方法的可行性。
Based on trait-marker multivariate regression, the method of quantitative trait loci （QTLs） mapping in each chromosome is proposed. A ‘pseudo-likelihood ratio’ statistic is used for significance test. Moreover, the detected QTLs are selected by the multivariate linear stepwise regression. This method is not only simplifying calculation, but also significantly increasing statistical power. A simulated example is given to elucidate this method effectively.
出处
《中国农学通报》
CSCD
2006年第7期130-134,共5页
Chinese Agricultural Science Bulletin
基金
福建省教育厅科技项目"线性模型及EM算法在QTL定位中的应用研究"(资助号JA04187)。
关键词
QTL定位
性状-标记回归
显著性检验
多元逐步回归
QTL mapping
trait-marker regression
significance test
multivariate linear stepwise regression
