For nearly two decades, genomic prediction and selection have supported efforts to increase genetic gains in plant and animal improvement programs. However, novel phenomic strategies for predicting complex traits in maize have recently proven beneficial when integrated into across‐environment sparse genomic prediction models. One phenomic data modality is whole grain near‐infrared spectroscopy (NIRS), which records reflectance values of biological samples (e.g., maize kernels) based on chemical composition. Predictions of hybrid maize grain yield (GY) and 500‐kernel weight (KW) across 2 years (2011–2012) and two management conditions (water‐stressed and well‐watered) were conducted using combinations of reflectance data obtained from high‐throughput, F 2 whole‐kernel scans and genomic data obtained from genotyping‐by‐sequencing within four different cross‐validation (CV) schemes (CV2, CV1, CV0, and CV00). When predicting the performance of untested genotypes in characterized (CV1) environments, genomic data were better than phenomic data for GY (0.689 ± 0.024—genomic vs. 0.612 ± 0.045—phenomic), but phenomic data were better than genomic data for KW (0.535 ± 0.034—genomic vs. 0.617 ± 0.145—phenomic). Multi‐kernel models (combinations of phenomic and genomic relationship matrices) did not surpass single‐kernel models for GY prediction in CV1 or CV00 (prediction of untested genotypes in uncharacterized environments); however, these models did outperform the single‐kernel models for prediction of KW in these same CVs. Lasso regression applied to the NIRS data set selected a subset of 216 NIRS bands that achieved comparable prediction abilities to the full phenomic data set of 3112 bands predicting GY and KW under CV1 and CV00.
The genetic yield potential in grain sorghum [Sorghum bicolor (L.) Moench] hybrids has increased at a slower rate than other cereal crops. While there are many reasons for this lag, increasing heterosis through broadening genetic distance between parents is commonly hypothesized as a mechanism to boost the rate of gain. To assess the relationship between heterotic groups in sorghum, a panel of 435 inbred lines from the Texas A&M program was genotyped using Genotyping-By-Sequencing (GBS). Lines were divided into three groups through K-means clustering based on genetic similarity. Distinct groups of B-lines, R-lines, and forage lines were observed, but they were not highly divergent from one another. In both elite germplasm and a large panel of sorghum lines, less diversity was observed among B-lines than R-lines. From the panel of 435 inbreds, a subset of 24 elite grain-type lines was chosen and crossed to produce 52 hybrids. Hybrids and parental lines were evaluated in four environments for grain yield, plant height, days to flower, and panicle exsertion. Genetic similarity among elite inbred parents of hybrids evaluated ranged from 0.632 to 0.792. For all traits, significant mid-parent heterosis was observed; however, correlations between mid-parent heterosis and genetic similarity were low to non-significant. GBS-based genetic similarity is not a good predictor of heterosis or hybrid performance among elite Texas A&M sorghum inbreds. However, some parental-inbred performance measurements may be predictive of hybrid performance.
Background: The structural characteristics of whole sorghum kernels are known to affect end-use quality, but traditional evaluation of this structure is two-dimensional (i.e., cross section of a kernel). Current technology offers the potential to consider three-dimensional structural characteristics of grain. X-ray computed tomography (CT) presents one such opportunity to nondestructively extract quantitative data from grain caryopses which can then be related to end-use quality. Results: Phenotypic measurements were extracted from CT scans of grain sorghum caryopses. Extensive phenotypic variation was found for embryo volume, endosperm hardness, endosperm texture, endosperm volume, pericarp volume, and kernel volume. CT derived estimates were strongly correlated with ground truth measurements enabling the identification of genotypes with superior structural characteristics. Conclusions: Presented herein is a phenotyping pipeline developed to quantify three-dimensional structural characteristics from grain sorghum caryopses which increases the throughput efficiency of previously difficult to measure traits. Adaptation of this workflow to other small-seeded crops is possible providing new and unique opportunities for scientists to study grain in a nondestructive manner which will ultimately lead to improvements end-use quality.
The post‐emergent herbicide options to control grasses in sorghum have historically been limited. Therefore, a mass selection program was implemented to increase the tolerance of sorghum to the broad‐spectrum herbicide tembotrione. The objectives of this study are to determine (i) which observation timing and vegetative indices are most effective in assessing sorghum injury to tembotrione using unoccupied aerial system imagery and (ii) if mass selection can be used to increase the tolerance of sorghum to tembotrione. Using an unoccupied aerial system, several vegetative indices, collected at either 14 or 21 days after tembotrione application, accurately measured sorghum injury. Over four cycles of selection, tembotrione tolerance increased in a linear and consistent manner. This demonstrates that mass selection can be used as a strategy for developing herbicide tolerance in crops. The modest, but steady genetic gain indicates both the quantitative nature of the trait and sufficient heritability to improve it. Given further development, tembotrione‐resistant sorghum hybrids could provide an effective means of post‐emergent weed control for a range of common weeds.