《Table 3–FSTvalues estimated between pairwise comparisons among three tested populations and among t

  提示：宽带有限、当前游客访问压缩模式

本系列图表出处文件名：随高清版一同展现

《Factors affecting genomic selection revealed by empirical evidence in maize》

1. 获取 高清版本忘记账户？点击这里登录

1. 下载图表忘记账户？点击这里登录

aNAT，natural population；RIL&DH，a population containing RILs and DH lines；F2:3，F3families derived from F2plants；TEM，temperate group；TRO，tropical group.

The estimates of rMGcontinuously increased as marker density and population size increased until they reached plateaux.For example，in the natural population，only a non-significant increase was obtained on rMGfor yield-related traits，when the number of SNPs increased from 7000 to 30，000，and the rMG estimate based on 37，803 markers was similar to that from 7000high-quality markers selected randomly（Fig.1a，c）.However，only a slight gain in rMGwas obtained when the number of markers increased from 1000 to 8000 in the RIL&DH population，showing that 1000 markers were enough to achieve a reasonable accuracy in biparental populations（Fig.1b，d）.As the population size increased，rMGincreased，and a slight gain in rMGwas obtained when the population size reached 250（Fig.2）.Generally，targeted traits with high h2showed high prediction accuracies.The correlation coefficients between rMGand heritability among traits were 0.83（P=0.04），0.89（P=0.02）and 0.95（P<0.01）in natural，RIL&DH and F2:3populations，respectively（Table 1）.Further improvement on prediction accuracy could be achieved with increased h2.