Significant QTL for sucrose concentration have been identified using a historical soybean genomic panel, which could aid in the development of food-grade soybean cultivars. Soybean (Glycine max (L.) Merr) is a crop of global importance for both human and animal consumption, which was domesticated in China more than 6000 years ago. A concern about losing genetic diversity as a result of decades of breeding has been expressed by soybean researchers. In order to develop new cultivars, it is critical for breeders to understand the genetic variability present for traits of interest in their program germplasm. Sucrose concentration is becoming an increasingly important trait for the production of soy-food products. The objective of this study was to use a genome-wide association study (GWAS) to identify putative QTL for sucrose concentration in soybean seed. A GWAS panel consisting of 266 historic and current soybean accessions was genotyped with 76 k genotype-by-sequencing (GBS) SNP data and phenotyped in four field locations in Ontario (Canada) from 2015 to 2017. Seven putative QTL were identified on chromosomes 1, 6, 8, 9, 10, 13 and 14. A key gene related to sucrose synthase (Glyma.06g182700) was found to be associated with the QTL located on chromosome 6. This information will facilitate efforts to increase the available genetic variability for sucrose concentration in soybean breeding programs and develop new and improved high-sucrose soybean cultivars suitable for the soy-food industry.

The accurate determination of soybean pubescence is essential for plant breeding programs and cultivar registration. Currently, soybean pubescence is classified visually, which is a labor-intensive and time-consuming activity. Additionally, the three classes of phenotypes (tawny, light tawny, and gray) may be difficult to visually distinguish, especially the light tawny class where misclassification with tawny frequently occurs. The objectives of this study were to solve both the throughput and accuracy issues in the plant breeding workflow, develop a set of indices for distinguishing pubescence classes, and test a machine learning (ML) classification approach. A principal component analysis (PCA) on hyperspectral soybean plot data identified clusters related to pubescence classes, while a Jeffries-Matusita distance analysis indicated that all bands were important for pubescence class separability. Aerial images from 2018, 2019, and 2020 were analyzed in this study. A 60-plot test (2019) of genotypes with known pubescence was used as reference data, while whole-field images from 2018, 2019, and 2020 were used to examine the broad applicability of the classification methodology. Two indices, a red/blue ratio and blue normalized difference vegetation index (blue NDVI), were effective at differentiating tawny and gray pubescence types in high-resolution imagery. A ML approach using a support vector machine (SVM) radial basis function (RBF) classifier was able to differentiate the gray and tawny types (83.1% accuracy and kappa = 0.740 on a pixel basis) on images where reference training data was present. The tested indices and ML model did not generalize across years to imagery that did not contain the reference training panel, indicating limitations of using aerial imagery for pubescence classification in some environmental conditions. High-throughput classification of gray and tawny pubescence types is possible using aerial imagery, but light tawny soybeans remain difficult to classify and may require training data from each field season.

Abstract Soybean (Glycine max (L.) Merr) is a crop of global importance for both human and animal consumption, which was domesticated in China more than 6000 years ago. A concern about losing genetic diversity as a result of decades of breeding has been expressed by soybean researchers. In order to develop new cultivars, it is critical for breeders to understand the genetic variability present for traits of interest in their program germplasm. Sucrose concentration is becoming an increasingly important trait for the production of soy-food products. The objective of this study was to use a genome-wide association study (GWAS) to identify putative QTL for sucrose concentration in soybean seed. A GWAS panel consisting of 266 historic and current soybean accessions was genotyped with 76k genotype-by-sequencing (GBS) SNP data and phenotyped in four field locations in Ontario (Canada) from 2015 to 2017. Seven putative QTL were identified on chromosomes 1, 6, 8, 9, 10, 13 and 14. A key gene related to sucrose synthase (Glyma.06g182700) was found to be associated with the QTL found on chromosome 6. This information will facilitate efforts to increase the available genetic variability for sucrose concentration in soybean breeding programs and develop new and improved high-sucrose soybean cultivars suitable for the soy-food industry.

Soybean [Glycine max (L.) Merr.] is the world's leading legume crop and the largest oilseed crop. It forms a symbiotic relationship with rhizobia bacteria residing in root nodules that provide fixed nitrogen to host plants through symbiotic nitrogen fixation (SNF). In soybean, it has been widely reported that the highest SNF occurs at the pod-filling stage, associated with the peak demand for nitrogen. However, the majority of seed nitrogen is derived from remobilizing root/shoot nitrogen, representing cumulative SNF from the seedling stage to the pre-pod-fill stage. Therefore, the question arises as to whether there has also been selection for improved SNF at these earlier stages, or whether pre-pod-fill SNF traits have drifted. To test this hypothesis, in this study, pre-pod SNF-related traits were evaluated in soybean cultivars that span 100 years of breeding selection in the Canadian Province of Ontario. Specifically, we evaluated SNF traits in 19 pedigree-related historical cultivars and 25 modern cultivars derived from the University of Guelph soybean breeding program. Field trials were conducted at Woodstock, Ontario, Canada in 2016 and 2017, and various SNF-related traits were measured at pre-pod-fill stages (R1-R3), including nitrogen fixation capacity. Considerable variation was observed among Canadian soybean cultivars released over the past 100 years for pre-pod-fill nitrogen fixation. The modern soybean cultivars had similar or moderately higher pre-pod-fill SNF compared to the historical lines in terms of the percentage of nitrogen derived from the atmosphere (%Ndfa) and total shoot fixed nitrogen. These findings suggest that, despite no direct selection by breeders, pre-pod-fill nitrogen fixation, and associated SNF traits have been maintained and possibly improved in modern soybean breeding. However, the low level of pre-pod-fill SNF in some modern cultivars, and generally wide variation observed in SNF between them, suggest some level of genetic drift for this trait in some pedigrees. Specific historical and modern soybean cultivars were identified as potential parents to enable targeted breeding for improved pre-pod-fill SNF. This retrospective study sheds light on our understanding of the impact of decades of recent selective breeding on pre-pod-fill nitrogen fixation traits in soybean in a temperate environment.

Identification of marker–trait associations and trait-associated haplotypes in breeding germplasm identifies regions under selection and highlights changes in haplotype diversity over decades of soybean improvement in Canada. Understanding marker–trait associations using genome-wide association in soybean is typically carried out in diverse germplasm groups where identified loci are often not applicable to soybean breeding efforts. To address this challenge, this study focuses on defining marker–trait associations in breeding germplasm and studying the underlying haplotypes in these regions to assess genetic change through decades of selection. Phenotype data were generated for 175 accessions across multiple environments in Ontario, Canada. A set of 76,549 SNPs were used in the association analysis. A total of 23 genomic regions were identified as significantly associated with yield (5), days to maturity (5), seed oil (3), seed protein (5) and 100-seed weight (5), of which 14 are novel. Each significant region was haplotyped to assess haplotype diversity of the underlying genomic region, identifying ten regions with trait-associated haplotypes in the breeding germplasm. The range of genomic length for these regions (7.2 kb to 6.8 Mb) indicates variation in regional LD for the trait-associated regions. Six of these regions showed changes between eras of breeding, from historical to modern and experimental soybean accessions. Continued selection on these regions may necessitate introgression of novel parental genetic diversity as some haplotypes were fixed within the breeding germplasm. This finding highlights the importance of studying associations and haplotype diversity at a breeding program scale to understand breeders’ selections and trends in soybean improvement over time. The haplotypes may also be used as a tool for selection of parental germplasm to inform breeder’s decisions on further soybean improvement.

The plant phenotyping community is adopting technological innovations in order to record phenotypic attributes more quickly and objectively. Low altitude aerial imaging is an appealing option for increasing throughput but there are still challenges in the image processing pipeline. One such challenge involves the assignment of a spatial reference to each plot entry in an experimental layout. Image‐based approaches are increasingly popular since plot boundaries are often, but not always, clearly visible in low altitude imagery. In addition, workflows that make geometric assumptions about plot layout also show promise. We outline an alternative approach to generate plot boundaries to overlay with aerial imagery. The proposed method involves high‐accuracy georeferencing (i.e., within a few cm) of imagery and planter activity, after which georeferencing of all plot entries is complete and only requires a few simple steps to convert logged spatial positions to polygons using open source geographic information systems (GIS) software. Compared with other approaches, the proposed method provides imagery that is precisely aligned over time and always aligns with plot boundaries, which are fixed and do not vary from image to image.

Improvement of agronomic and seed quality traits in soybean [Glycine max (L.) Merr.] occurs in breeding programs as a result of crossing elite soybean genotypes and selection of superior offspring in each cycle. The objective of this study was to evaluate changes in traits over 100 yr of selection in Ontario, Canada, in two soybean breeding programs at the University of Guelph: Guelph Campus (Maturity Group [MG] 0 and I) and Ridgetown Campus (MG II) programs. Fifty‐six Guelph cultivars (released between 1983 and 2016), 46 Ridgetown cultivars (released between 1987 and 2015), and 37 Guelph‐pedigree related historical cultivars (from 1913 to 1985) were grown in field trials over 3 yr (Guelph Campus) and 2 yr (Ridgetown Campus) at two locations per breeding program to measure 13 traits. Yield increased in the Guelph Campus cultivars by 17.1 kg ha−1 yr−1 and by 15.7 kg ha−1 yr−1 in the Ridgetown Campus cultivars, while yield per day of maturity per year of release significantly increased in elite cultivars. Protein in the Ridgetown Campus cultivars increased at 0.76 g kg−1 yr−1, whereas the Guelph Campus cultivars showed no trend. In Ridgetown Campus cultivars 100‐seed weight significantly increased at 0.1 g yr−1. Within the historical cultivars, 100‐seed weight was significantly correlated with both protein (r = 0.50) and oil (r = −0.42). Genotype × trait biplot analyses revealed differential patterns of trait relationships between the elite and historical cultivars. Oil and yield were associated within the Guelph Campus cultivars. The association of 100‐seed weight and protein was only found in the historical cultivars. Yield has increased in both breeding programs while seed traits have been altered over time.

Genetic diversity in Canadian soybean is maintained over decades of selection in two public breeding programs. Breeders have used a portion of the genetic diversity available in germplasm collections. Both public and private breeding efforts have been critical for the development of soybean cultivars grown around the world. Global genetic diversity of soybean has been well characterized; however, this diversity is not well studied at the breeding program scale. The objective of this study was to characterize genetic diversity over decades of breeding in two public soybean breeding programs at the University of Guelph, Canada. To address this objective, a pedigree-related panel combining 296 soybean accessions from the Ridgetown and Guelph Campus breeding programs was studied. The accessions were genotyped using genotyping-by-sequencing, imputed using the GmHapMap reference genotypes resulting in more than 3.8M SNPs, further filtered to 77k SNPs. Population structure analysis did not identify structure between the breeding programs and historical germplasm. The linkage disequilibrium decay ranged from 400 to 600 kb on average in euchromatic regions. Nucleotide diversity over decades of breeding shows that historical accessions had the highest nucleotide diversity, with significant decreases corresponding to the initial breeding activity in Canada; however, genetic diversity has increased in the last 20 years in both breeding programs. Maturity gene E2 was nearly fixed for e2 in Ridgetown accessions, while unfixed in Guelph accessions. Comparison of the breeding programs to the USDA germplasm collection reveals that breeders have only used a portion of the available genetic diversity, allowing future breeders to exploit this untapped resource. The approach used in this study may be of interest to other breeding programs for evaluating changes in genetic diversity resulting from breeding activities.

NSERC Collaborative Research and Development Grant, Huron Commodities, SeCan, Grain Farmers of Ontario