Establishment of a quasi-field trial in Abies nordmanniana-test of a new approach to forest tree breeding
Establishment of a quasi-field trial in Abies nordmanniana-test of a new approach to forest tree breeding
复制标题
DOI：
10.1007/s11295-009-0253-6
复制
发表时间：
2010-02-01
影响因子：
2.4
通讯作者：
McKinney, Lea V.
中科院分区：
文献类型：
--
作者：
Hansen, Ole K.;McKinney, Lea V.
这项研究使用 DNA 标记在生产圣诞树台内建立了准田间试验，该圣诞树台是由在开放授粉克隆种子果园 (CSO) 收集的种子生产的。 CSO 共有 660 个后代，其中包括 99 个冷杉克隆，用 12 个微卫星进行了基因分型。 93% 和 98% 的后代成功分配亲子关系，置信度分别为 95% 和 80%。当标记数量减少到 10 时，分配率仅下降至 90%。CSO 克隆中后代的亲本分布高度倾斜。在 7% 的情况下，最成功的克隆被指定为亲本，而 119 个潜在亲本基因型中只有 92 个被指定为亲本。获得的谱系用于估计与圣诞树育种相关的五个性状的 CSO 克隆的育种值。对于高遗传力性状，例如潮红，可以估计相当大比例的克隆的准确育种值。为了估计低遗传力性状的育种值，例如圣诞树质量评分，将需要更多的基因分型后代。该方法的最大缺点是种子园中亲本的亲本分布高度倾斜，使得计算所有克隆的育种值变得困难。该方法似乎非常适合树木育种，它更强调亲本基因型的纯粹选择，而不是估计定量遗传参数。
This study used DNA markers to establish a quasi-field trial within a production Christmas tree stand produced from seed collected in an open-pollinated clonal seed orchard (CSO). A total of 660 offspring from the CSO, which comprised 99 clones of Abies nordmanniana, were genotyped with 12 microsatellites. Parentage was assigned successfully to 93% and 98% of the progeny at 95% and 80% confidence, respectively. The assignment rate declined only to 90% when the number of markers was reduced to 10. The distribution of parentage to the offspring among the CSO clones was highly skewed. The most successful clone was assigned as parent in 7% of the cases, and only 92 of the 119 potential parental genotypes were assigned as parents. The obtained pedigree was used to estimate breeding values for the CSO clones for five characters relevant for Christmas tree breeding. For high-heritability traits, such as flushing, accurate breeding values could be estimated for a considerable proportion of the clones. To estimate breeding values for low-heritability traits, such as Christmas tree quality score, more genotyped offspring will be required. The largest drawback of the method is the highly skewed distribution of parentage among the parents in the seed orchards, making it difficult to calculate breeding values for all clones. The approach seems well suited for tree breeding that puts more emphasis on pure selection of parental genotypes and less on estimating quantitative genetic parameters.