Legume Genomics and Genetics 2024, Vol.15, No.6, 270-279 http://cropscipublisher.com/index.php/lgg 272 3.2 Key findings from GWAS studies in soybean Several GWAS studies have identified key loci associated with important agronomic traits in soybean. For instance, a study identified significant SNPs associated with maturity, plant height, seed weight, and yield, with some SNPs mapped to known loci such as E2, E4, and Dt1 (Ravelombola et al., 2021). Another study focusing on soybean germplasm derived from Canadian × Chinese crosses identified QTL regions controlling seed yield, oil, and protein content, highlighting the potential of Chinese cultivars in improving these traits (Priyanatha et al., 2022). Additionally, a haplotype-based GWAS identified stable haplotype associations for seed yield and seed weight across different environments, providing insights into the genetic determinants of these traits (Contreras-Soto et al., 2017). 3.3 Challenges and limitations of using GWAS in soybean breeding Despite its advantages, GWAS in soybean breeding faces several challenges. One major limitation is the insufficient statistical power to detect QTL with small effects, especially in populations with narrow genetic bases (Yoosefzadeh-Najafabadi et al., 2021a). Additionally, the complex genetic architecture of traits, involving multiple loci with small effects and gene-environment interactions, complicates the identification of significant associations (Yoosefzadeh-Najafabadi et al., 2023). Another challenge is the need for high-density SNP markers and large, diverse populations to ensure robust and reproducible results (Sonah et al., 2015; Mandozai et al., 2021). Furthermore, integrating GWAS findings into practical breeding programs requires validation of identified markers and candidate genes, which can be resource-intensive and time-consuming. 4 Advances in Genomic Selection (GS) 4.1 Overview of gs and its application in crop improvement Genomic Selection (GS) is a revolutionary approach in plant breeding that leverages genome-wide marker data to predict the breeding values of individuals within a population. Unlike traditional marker-assisted selection, which focuses on a few significant markers, GS incorporates all available marker information into the prediction model. This comprehensive approach allows for the capture of small-effect Quantitative Trait Loci (QTL) that contribute to complex traits, thereby improving the accuracy of selection and accelerating the breeding cycle (Jannink et al., 2010; Varshney et al., 2017; Merrick et al., 2022). GS has been successfully applied in various crop improvement programs, including those for cereals like wheat, maize, and rice, as well as legumes such as soybeans. The method has shown promise in enhancing traits related to yield, quality, and stress tolerance, which are often controlled by multiple genes with small effects (Krishnappa et al., 2021; Xu et al., 2021; Budhlakoti et al., 2022). By integrating GS with other advanced technologies like high-throughput phenotyping and deep learning, breeding programs can achieve more rapid and cost-effective genetic gains (Krishnappa et al., 2021; Merrick et al., 2022). 4.2 Differences between gs and traditional marker-assisted selection Traditional Marker-Assisted Selection (MAS) has been effective for traits controlled by a few major QTLs but has limitations when dealing with polygenic traits. MAS typically involves identifying and selecting for specific markers associated with large-effect QTLs, which can be a time-consuming and less efficient process for complex traits (Jannink et al., 2010; Varshney et al., 2017; Merrick et al., 2022). In contrast, GS uses genome-wide markers to estimate the effects of all loci simultaneously, providing a more holistic and accurate prediction of an individual's genetic potential. This method avoids the bias associated with selecting only significant markers and captures the cumulative effect of numerous small-effect QTLs. As a result, GS can accelerate the breeding cycle by allowing for earlier selection decisions based on Genomic Estimated Breeding Values (GEBVs) rather than waiting for phenotypic data (Wang et al., 2018). Additionally, GS can reduce the need for extensive phenotyping, thereby saving time and resources (Xu et al., 2021; Budhlakoti et al., 2022). 4.3 Successful cases of GS implementation in soybean breeding programs Several soybean breeding programs have successfully implemented GS to enhance the selection of desirable traits. For instance, a study involving 483 elite soybean breeding lines demonstrated the potential of GS for improving
RkJQdWJsaXNoZXIy MjQ4ODYzNA==