Predictive Characterization for Seed Morphometric Traits for Genebank Accessions Using Genomic Selection
Predictive Characterization for Seed Morphometric Traits for Genebank Accessions Using Genomic Selection
复制标题
DOI：
10.3389/fevo.2020.00032
复制
发表时间：
2020-03-17
影响因子：
3
通讯作者：
Amril, Ahmed
中科院分区：
文献类型：
--
作者：
Kehel, Zakaria;Sanchez-Garcia, Miguel;Amril, Ahmed
面包小麦的种子性状，包括被认为与作物早期活力和最终使用质量有关的种子大小，对农民和育种者很有价值。在这项研究中，使用GrainScan对789个面包小麦地方品种的种子形态特征进行了扫描，这些品种保存在国际干旱地区农业研究中心（ICARDA）的基因库中。利用12k个DartSeq SNP标记进行的多样性分析显示，这些材料可以分为五个不同的聚类。为了评估从基因库中早期选择的表现，我们检查了这些地方品种的基因组关系的基因组选择模型的准确性。交叉验证表明，种子周长和种子宽度的预测精度在0.64 ~ 0.74之间。随机验证预测准确度的变异平均为0.14，范围从0.12到0.18，表明即使从基因库中收集到更大的遗传多样性，结果也具有稳定的可预测性和可重复性。加入基于护照信息的气候关系矩阵，预测能力提高了8%。我们对种子性状的研究结果表明，可以直接根据基因组信息估计基因库中重要的农艺表型，进一步倡导基因组技术的进步，以确定亲本种质作为有益等位基因的潜在供体。
Seed traits of bread wheat, including the seed size that is considered to be associated with early vigor of the crop and end-use quality, are valuable to farmers and breeders. In this study, a collection of 789 bread wheat landraces, held in-trust at the genebank of the International Center for Agricultural Research in the Dry Areas (ICARDA) were scanned for seed morphometric traits using GrainScan. Diversity analysis using the 12k DartSeq SNP markers revealed that these accessions can be grouped into five distinct clusters. To evaluate the performance for early selection from genebank accessions, we examined the accuracy of genomic selection models with genomic relationship that these landraces accounted for. Based on cross-validations, prediction accuracies for seed traits ranged from 0.64 for seed perimeter to 0.74 for seed width. The variability of prediction accuracies across random validations averaged at 0.14, with a range from 0.12 to 0.18, suggesting stable predictability and reproducible results even with a collection of much greater genetic diversity from genebank accessions. Adding the climatic relationship matrix between accessions based on passport information improved the predictive ability by 8%. Our results on seed traits demonstrated the capacity for estimating important agronomic phenotypes for genebank accessions directly based on genomic information, further advocating the advance in genomic technologies for identifying parental germplasm as potential donors of beneficial alleles for introgression.