Meuwissen Theo H E, Goddard Mike E
Centre for Integrative Genetics (Cigene), Institute of Animal Science, Agricultural University of Norway, Box 5025, As, Norway.
Genet Sel Evol. 2004 May-Jun;36(3):261-79. doi: 10.1186/1297-9686-36-3-261.
A multi-locus QTL mapping method is presented, which combines linkage and linkage disequilibrium (LD) information and uses multitrait data. The method assumed a putative QTL at the midpoint of each marker bracket. Whether the putative QTL had an effect or not was sampled using Markov chain Monte Carlo (MCMC) methods. The method was tested in dairy cattle data on chromosome 14 where the DGAT1 gene was known to be segregating. The DGAT1 gene was mapped to a region of 0.04 cM, and the effects of the gene were accurately estimated. The fitting of multiple QTL gave a much sharper indication of the QTL position than a single QTL model using multitrait data, probably because the multi-locus QTL mapping reduced the carry over effect of the large DGAT1 gene to adjacent putative QTL positions. This suggests that the method could detect secondary QTL that would, in single point analyses, remain hidden under the broad peak of the dominant QTL. However, no indications for a second QTL affecting dairy traits were found on chromosome 14.
本文提出了一种多位点QTL定位方法,该方法结合了连锁和连锁不平衡(LD)信息,并使用多性状数据。该方法假定在每个标记区间的中点存在一个假定的QTL。使用马尔可夫链蒙特卡罗(MCMC)方法对假定的QTL是否有效应进行抽样。该方法在已知DGAT1基因在第14号染色体上分离的奶牛数据中进行了测试。DGAT1基因被定位到一个0.04 cM的区域,并且该基因的效应被准确估计。与使用多性状数据的单QTL模型相比,多个QTL的拟合给出了QTL位置更清晰的指示,这可能是因为多位点QTL定位减少了大的DGAT1基因对相邻假定QTL位置的残留效应。这表明该方法可以检测到在单点分析中会隐藏在显性QTL宽峰下的次要QTL。然而,在第14号染色体上未发现影响奶牛性状的第二个QTL的迹象。