【摘要】 通过比较矩估计法和最大似然法在标记－ＱＴＬ连锁检测、重组率估计及ＱＴＬ基因型的均值和方差估计方面的效率，探讨了样本容量对这些参数估计的影响。结果表明，ＬＯＤ估计值随作图群体样本容量的增加而增加，重组率估值的偏倚随样本容量的增加而减少。ＱＴＬ基因型均值和方差估计值偏倚的程度均随理论重组率的增加而增加，随群体样本容量的增加而减少，最大似然法的估计效率高于矩估计法，且当理论重组率较高时尤为明显，在ｒ＝０．０－０．３０范围内，要获得无偏的重组率、均值和方差估计值，采用矩估计法时，要求样本容量ｎ＞１０００，而采用最大似然法一般只要求样本容量，ｎ≥３００．当标记－ＱＴＬ完全独立时，两种方法均不能获得这些参数的无偏估计值。更多还原

【Abstract】 Effect of population size on the deteetion of linkage between a genetic marker and a quantitative trait locus(QTL)and the estimation of recombinant fraction,and means and vari-ances of QTL genotypes was studied by comparison of two methods,moment estimation(ME)and rnaxirnum likelihood estimation(MLE).The results indicate that estimators of LOD increase and deviation of estimated recombinant fraction from the expected decrease,when sample size increas-es. Deviation of estimators of means and variances of QTL genotypes increase with higher recombi-nant fraction and decrease with larger sample size.MLE obtained more accurate extimators than ME,particularly for higher expected recombinant fraction. In order to obtain unbiased estimators of recombinant fraction,means and variances,Sampled population size should larger than 1 000 for ME and 300 or more for MLE,if the expected recombinant fractions are within 0-0.30.When the marker is not linked with the putatlve QTL,neither of the methods can obtain the unbiased es-timators of these parameters.更多还原