Independent and Joint-GWAS for growth traits in Eucalyptus by assembling genome-wide data for 3373 individuals across four breeding populations

First author: Bárbara S. F. Müller; Affiliations: University of Brasília (巴西利亚大学): Brasília, Brazil

Corresponding author: Dario Grattapaglia 

Genome‐wide association studies (GWAS) in plants typically suffer from limited statistical power. An alternative to the logistical and cost challenge of increasing sample sizes is to gain power by meta‐analysis using information from independent studies. We carried out GWAS for growth traits with six single‐marker models and regional heritability mapping (RHM) in four Eucalyptus breeding populations independently and by Joint‐GWAS, using gene and segment‐ased models, with data for 3373 individuals genotyped with a communal EUChip60KSNP platform. While single‐single nucleotide polymorphism (SNP) GWAS hardly detected significant associations at high‐stringency (高严格性) in each population, gene‐ased Joint‐GWAS revealed nine genes significantly associated with tree height. Associations detected using single‐SNP GWAS, RHM and Joint‐GWAS set‐ased models explained on average 3-20% of the phenotypic variance. Whole‐genome regression (全基因组回归), conversely, captured 64-89% of the pedigree‐ased heritability in all populations. Several associations independently detected for the same SNPs in different populations provided unprecedented (前所未有的) GWAS validation results in forest trees. Rare and common associations were discovered in eight genes involved in cell wall biosynthesis and lignification. With the increasing adoption of genomic prediction of complex phenotypes using shared SNPs and much larger tree breeding populations, Joint‐GWAS approaches should provide increasing power to pinpoint (精准定位) discrete (离散型) associations potentially useful toward tree breeding and molecular applications.

植物中的全基因组关联分析GWAS严重受限于统计学。通过对不同独立研究数据的元分析可帮助解决样本越来越多所带来的挑战。作者通过6个单标记模型和区域遗传力作图RHM对四个桉树育种群体的生长性状进行GWAS分析，并利用基于基因和片段的模型对3373个商业化EUChip60KSNP平台的个体进行了Joint‐GWAS分析。在每个群体中，单个SNP GWAS几乎难以测试显著的关联，而基于基因的Joint‐GWAS坚定了9个显著与树高关联的基因。利用单个SNP GWAS、RHM和Joint‐GWAS鉴定的关联大约解释了表型变异的3-20%。全基因组回归大概在所有群体中捕获了大约64-89%的谱系遗传力。在不同群体中完全独立的关联分析鉴定到了相同的SNP，这在林木树种中相当于提供了非常坚定的GWAS验证结果。作者还在8个参与细胞壁合成及木质化的基因上鉴定到了罕见和普通的关联位点。随着更多的共享SNP复杂表型的基因组预测利用及更大的育种群体，Joint‐GWAS方法可能会更加适合用于对离散型关联位点的精确定位，这些位点的鉴定将有助于林木育种和分子应用。

通讯：Dario Grattapaglia（https://cnr.ncsu.edu/directory/dario-grattapaglia/）

个人简介：1985年，巴西利亚大学，林业工程学士；1994年，北卡罗来纳州立大学，遗传学与林业博士。

doi: https://doi.org/10.1111/nph.15449

Journal: New Phytologist

First Published: 25 September, 2018

（P.S. 原文下载：链接:https://pan.baidu.com/s/1qFiSsqwlF_7JjmGMbedfhQ  密码:dmu6）