Dissection of the practical soybean breeding pipeline by developing ZDX1, a high-throughput functional array

Jing Wen,Rongqi Guo,Bingfu Guo,Bincheng Sun,Yanfei Li,Qiang Li,Huilong Hong,Jiqiang Liu,Lili Yu,Jiangyuan Xu,Xindong Yao,Peng Yu ,Jun Wang,Lei Feng,Shen Chai,Xingguo Hu,Rongli Sun ,Shi Su ,Yu Tian,Yong Zhang,Qí Zhāng ,Erhu Su,Jingshun Wang,Yinghui Li,Lijuan Qiu,Huawei Gao,Ying Gu ,Yue Ma ,Wanhai Zhang,Huihui Li

doi:10.1007/s00122-022-04043-w

Abstract

Key messageWe developed the ZDX1 high-throughput functional soybean array for high accuracy evaluation and selection of both parents and progeny, which can greatly accelerate soybean breeding.Microarray technology facilitates rapid, accurate, and economical genotyping. Here, using resequencing data from 2214 representative soybean accessions, we developed the high-throughput functional array ZDX1, containing 158,959 SNPs, covering 90.92% of soybean genes and sites related to important traits. By application of the array, a total of 817 accessions were genotyped, including three subpopulations of candidate parental lines, parental lines and their progeny from practical breeding. The fixed SNPs were identified in progeny, indicating artificial selection during the breeding process. By identifying functional sites of target traits, novel soybean cyst nematode-resistant progeny and maturity-related novel sources were identified by allele combinations, demonstrating that functional sites provide an efficient method for the rapid screening of desirable traits or gene sources. Notably, we found that the breeding index (BI) was a good indicator for progeny selection. Superior progeny were derived from the combination of distantly related parents, with at least one parent having a higher BI. Furthermore, new combinations based on good performance were proposed for further breeding after excluding redundant and closely related parents. Genomic best linear unbiased prediction (GBLUP) analysis was the best analysis method and achieved the highest accuracy in predicting four traits when comparing SNPs in genic regions rather than whole genomic or intergenic SNPs. The prediction accuracy was improved by 32.1% by using progeny to expand the training population. Collectively, a versatile assay demonstrated that the functional ZDX1 array provided efficient information for the design and optimization of a breeding pipeline for accelerated soybean breeding.

Highlights

The goal of crop breeding is to develop plant varieties with ideal traits, such as higher yield, improved quality, and enhanced environmental adaptability
The single nucleotide polymorphisms (SNPs) selected for inclusion in the ZDX1 array included 14,685 synonymous sites, 6120 unknown sites, 14,845 sites located in intronic regions, 12,158 sites located within 1000 bp upstream or downstream of a gene, 9804 sites located in untranslated regions, and 94,524 sites located in intergenic regions (Supplemental Table 1)
The previous soybean arrays were developed based on information obtained from only a few to dozens of cultivated or wild species (Lee et al 2015; Song et al 2013; Wang et al 2016)

Summary

Introduction

The goal of crop breeding is to develop plant varieties with ideal traits, such as higher yield, improved quality, and enhanced environmental adaptability. The 50 K soybean array, which has higher density, was used to genotype 96 elite, landrace, and wild accessions, and to identify candidate genomic regions shaped by domestication or recent selection (Song et al 2013). This array was used to correlate protein- and oil-related loci via genome-wide association study (GWAS) analysis of 298 strains (Hwang et al 2014). When 180 K (Lee et al 2015) and 355 K (Wang et al 2016) arrays were developed, natural hybrids between cultivated and wild soybean, as well as a candidate interval affecting grain weight, were identified These arrays laid the foundation for the application of SNP arrays in genetic research and molecular breeding

Methods

Results

Conclusion