Optimizing feature selection with gradient boosting machines in PLS regression for predicting moisture and protein in multi-country corn kernels via NIR spectroscopy.

Differences in moisture and protein content impact both nutritional value and processing efficiency of corn kernels. Near-infrared (NIR) spectroscopy can be used to estimate kernel composition, but models trained on a few environments may underestimate error rates and bias. We assembled corn samples from diverse international environments and used NIR with chemometrics and partial least squares regression (PLSR) to determine moisture and protein. The potential of five feature selection methods to improve prediction accuracy was assessed by extracting sensitive wavelengths. Gradient boosting machines (GBMs), particularly CatBoost and LightGBM, were found to effectively select crucial wavelengths for moisture (1409, 1900, 1908, 1932, 1953, 2174 nm) and protein (887, 1212, 1705, 1891, 2097, 2456 nm). SHAP plots highlighted significant wavelength contributions to model prediction. These results illustrate GBMs' effectiveness in feature engineering for agricultural and food sector applications, including developing multi-country global calibration models for moisture and protein in corn kernels.

Rhythmic lipid and gene expression responses to chilling in panicoid grasses.

Chilling stress threatens plant growth and development, particularly affecting membrane fluidity and cellular integrity. Understanding plant membrane responses to chilling stress is important for unraveling the molecular mechanisms of stress tolerance. Whereas core transcriptional responses to chilling stress and stress tolerance are conserved across species, the associated changes in membrane lipids appear to be less conserved, as which lipids are affected by chilling stress varies by species. Here, we investigated changes in gene expression and membrane lipids in response to chilling stress during one 24 hour cycle in chilling-tolerant foxtail millet (Setaria italica), and chilling-sensitive sorghum (Sorghum bicolor), and Urochloa (browntop signal grass, Urochloa fusca, lipids only), leveraging their evolutionary relatedness and differing levels of chilling-stress tolerance. We show that most chilling-induced lipid changes are conserved across the three species, while we observed distinct, time-specific responses in chilling-tolerant foxtail millet, indicating the presence of a finely orchestrated adaptive mechanism. We detected rhythmicity in lipid responses to chilling stress in the three grasses, which were also present in Arabidopsis (Arabidopsis thaliana), suggesting the conservation of rhythmic patterns across species and highlighting the importance of accounting for time of day. When integrating lipid datasets with gene expression profiles, we identified potential candidate genes that showed corresponding transcriptional changes in response to chilling stress, providing insights into the differences in regulatory mechanisms between chilling-sensitive sorghum and chilling-tolerant foxtail millet.

Population-level gene expression can repeatedly link genes to functions in maize.

Transcriptome-wide association studies (TWAS) can provide single gene resolution for candidate genes in plants, complementing genome-wide association studies (GWAS) but efforts in plants have been met with, at best, mixed success. We generated expression data from 693 maize genotypes, measured in a common field experiment, sampled over a 2-h period to minimize diurnal and environmental effects, using full-length RNA-seq to maximize the accurate estimation of transcript abundance. TWAS could identify roughly 10 times as many genes likely to play a role in flowering time regulation as GWAS conducted data from the same experiment. TWAS using mature leaf tissue identified known true-positive flowering time genes known to act in the shoot apical meristem, and trait data from a new environment enabled the identification of additional flowering time genes without the need for new expression data. eQTL analysis of TWAS-tagged genes identified at least one additional known maize flowering time gene through trans-eQTL interactions. Collectively these results suggest the gene expression resource described here can link genes to functions across different plant phenotypes expressed in a range of tissues and scored in different experiments.

Genotype-specific nonphotochemical quenching responses to nitrogen deficit are linked to chlorophyll a to b ratios.

Non-photochemical quenching (NPQ) protects plants from photodamage caused by excess light energy. Substantial variation in NPQ has been reported among different genotypes of the same species. However, comparatively little is known about how environmental perturbations, including nutrient deficits, impact natural variation in NPQ kinetics. Here, we analyzed a natural variation in NPQ kinetics of a diversity panel of 225 maize (Zea mays L.) genotypes under nitrogen replete and nitrogen deficient field conditions. Individual maize genotypes from a diversity panel exhibited a range of changes in NPQ in response to low nitrogen. Replicated genotypes exhibited consistent responses across two field experiments conducted in different years. At the seedling and pre-flowering stages, a similar portion of the genotypes (∼33%) showed decrease, no-change or increase in NPQ under low nitrogen relative to control. Genotypes with increased NPQ under low nitrogen also showed greater reductions in dry biomass and photosynthesis than genotypes with stable NPQ when exposed to low nitrogen conditions. Maize genotypes where an increase in NPQ was observed under low nitrogen also exhibited a reduction in the ratio of chlorophyll a to chlorophyll b. Our results underline that since thermal dissipation of excess excitation energy measured via NPQ helps to balance the energy absorbed with energy utilized, the NPQ changes are the reflection of broader molecular and biochemical changes which occur under the stresses such as low soil fertility. Here, we have demonstrated that variation in NPQ kinetics resulted from genetic and environmental factors, are not independent of each other. Natural genetic variation controlling plastic responses of NPQ kinetics to environmental perturbation increases the likelihood it will be possible to optimize NPQ kinetics in crop plants for different environments.

Image Filtering to Improve Maize Tassel Detection Accuracy Using Machine Learning Algorithms.

Unmanned aerial vehicle (UAV)-based imagery has become widely used to collect time-series agronomic data, which are then incorporated into plant breeding programs to enhance crop improvements. To make efficient analysis possible, in this study, by leveraging an aerial photography dataset for a field trial of 233 different inbred lines from the maize diversity panel, we developed machine learning methods for obtaining automated tassel counts at the plot level. We employed both an object-based counting-by-detection (CBD) approach and a density-based counting-by-regression (CBR) approach. Using an image segmentation method that removes most of the pixels not associated with the plant tassels, the results showed a dramatic improvement in the accuracy of object-based (CBD) detection, with the cross-validation prediction accuracy (r2) peaking at 0.7033 on a detector trained with images with a filter threshold of 90. The CBR approach showed the greatest accuracy when using unfiltered images, with a mean absolute error (MAE) of 7.99. However, when using bootstrapping, images filtered at a threshold of 90 showed a slightly better MAE (8.65) than the unfiltered images (8.90). These methods will allow for accurate estimates of flowering-related traits and help to make breeding decisions for crop improvement.

Variation in Leaf Chlorophyll Concentration in Response to Nitrogen Application Across Maize Hybrids in Contrasting Environments.

Leaf chlorophyll concentration was measured for 84 publicly available maize hybrids grown under three nitrogen fertilizer treatments in two contrasting environments in Nebraska. The effect of nitrogen treatment on chlorophyll response was found to be significant (p < 0.05) for both locations. In Scottsbluff, chlorophyll concentrations increased significantly with increasing nitrogen rate, while no significant difference was found between medium and high nitrogen in Lincoln. Within equivalent nitrogen treatments, chlorophyll was more abundant in Lincoln than Scottsbluff for nearly every hybrid. Hybrid response was not consistent between environments, with approximately 11% of variance explained by genotype by environment interaction.

A Transferable Machine Learning Framework for Predicting Transcriptional Responses of Genes Across Species.

Leveraging existing resources in studied species to predict gene functions has the potential to rapidly expand understanding of annotated genes in other, less well-studied, species with assembled genomes. However, orthology is not a reliable predictor for the transcriptional responses of genes to stress. Machine learning methods can quantitatively estimate expression patterns and gene functions using known annotations and collections of features describing each gene. In this chapter, we describe a supervised machine learning framework to predict stress-responsive genes across species using only features derived from nucleotide sequences, using the example of cold stress-responsive genes in different Panicoid grass species.

Genomes to Fields 2022 Maize genotype by Environment Prediction Competition.

OBJECTIVES: The Genomes to Fields (G2F) 2022 Maize Genotype by Environment (GxE) Prediction Competition aimed to develop models for predicting grain yield for the 2022 Maize GxE project field trials, leveraging the datasets previously generated by this project and other publicly available data. DATA DESCRIPTION: This resource used data from the Maize GxE project within the G2F Initiative [1]. The dataset included phenotypic and genotypic data of the hybrids evaluated in 45 locations from 2014 to 2022. Also, soil, weather, environmental covariates data and metadata information for all environments (combination of year and location). Competitors also had access to ReadMe files which described all the files provided. The Maize GxE is a collaborative project and all the data generated becomes publicly available [2]. The dataset used in the 2022 Prediction Competition was curated and lightly filtered for quality and to ensure naming uniformity across years.

A complete telomere-to-telomere assembly of the maize genome.

A complete telomere-to-telomere (T2T) finished genome has been the long pursuit of genomic research. Through generating deep coverage ultralong Oxford Nanopore Technology (ONT) and PacBio HiFi reads, we report here a complete genome assembly of maize with each chromosome entirely traversed in a single contig. The 2,178.6 Mb T2T Mo17 genome with a base accuracy of over 99.99% unveiled the structural features of all repetitive regions of the genome. There were several super-long simple-sequence-repeat arrays having consecutive thymine-adenine-guanine (TAG) tri-nucleotide repeats up to 235 kb. The assembly of the entire nucleolar organizer region of the 26.8 Mb array with 2,974 45S rDNA copies revealed the enormously complex patterns of rDNA duplications and transposon insertions. Additionally, complete assemblies of all ten centromeres enabled us to precisely dissect the repeat compositions of both CentC-rich and CentC-poor centromeres. The complete Mo17 genome represents a major step forward in understanding the complexity of the highly recalcitrant repetitive regions of higher plant genomes.

Oligogalactolipid production during cold challenge is conserved in early diverging lineages.

Severe cold, defined as a damaging cold beyond acclimation temperatures, has unique responses, but the signaling and evolution of these responses are not well understood. Production of oligogalactolipids, which is triggered by cytosolic acidification in Arabidopsis (Arabidopsis thaliana), contributes to survival in severe cold. Here, we investigated oligogalactolipid production in species from bryophytes to angiosperms. Production of oligogalactolipids differed within each clade, suggesting multiple evolutionary origins of severe cold tolerance. We also observed greater oligogalactolipid production in control samples than in temperature-challenged samples of some species. Further examination of representative species revealed a tight association between temperature, damage, and oligogalactolipid production that scaled with the cold tolerance of each species. Based on oligogalactolipid production and transcript changes, multiple angiosperm species share a signal of oligogalactolipid production initially described in Arabidopsis, namely cytosolic acidification. Together, these data suggest that oligogalactolipid production is a severe cold response that originated from an ancestral damage response that remains in many land plant lineages and that cytosolic acidification may be a common signaling mechanism for its activation.

Genetic control of photoprotection and photosystem II operating efficiency in plants.

Photoprotection against excess light via nonphotochemical quenching (NPQ) is indispensable for plant survival. However, slow NPQ relaxation under low light conditions can decrease yield of field-grown crops up to 40%. Using semi-high-throughput assay, we quantified the kinetics of NPQ and photosystem II operating efficiency (ΦPSII) in a replicated field trial of more than 700 maize (Zea mays) genotypes across 2 yr. Parametrized kinetics data were used to conduct genome-wide association studies. For six candidate genes involved in NPQ and ΦPSII kinetics in maize the loss of function alleles of orthologous genes in Arabidopsis (Arabidopsis thaliana) were characterized: two thioredoxin genes, and genes encoding a transporter in the chloroplast envelope, an initiator of chloroplast movement, a putative regulator of cell elongation and stomatal patterning, and a protein involved in plant energy homeostasis. Since maize and Arabidopsis are distantly related, we propose that genes involved in photoprotection and PSII function are conserved across vascular plants. The genes and naturally occurring functional alleles identified here considerably expand the toolbox to achieving a sustainable increase in crop productivity.

2018-2019 field seasons of the Maize Genomes to Fields (G2F) G x E project.

OBJECTIVES: This report provides information about the public release of the 2018-2019 Maize G X E project of the Genomes to Fields (G2F) Initiative datasets. G2F is an umbrella initiative that evaluates maize hybrids and inbred lines across multiple environments and makes available phenotypic, genotypic, environmental, and metadata information. The initiative understands the necessity to characterize and deploy public sources of genetic diversity to face the challenges for more sustainable agriculture in the context of variable environmental conditions. DATA DESCRIPTION: Datasets include phenotypic, climatic, and soil measurements, metadata information, and inbred genotypic information for each combination of location and year. Collaborators in the G2F initiative collected data for each location and year; members of the group responsible for coordination and data processing combined all the collected information and removed obvious erroneous data. The collaborators received the data before the DOI release to verify and declare that the data generated in their own locations was accurate. ReadMe and description files are available for each dataset. Previous years of evaluation are already publicly available, with common hybrids present to connect across all locations and years evaluated since this project's inception.

A leaf-level spectral library to support high-throughput plant phenotyping: predictive accuracy and model transfer.

Leaf-level hyperspectral reflectance has become an effective tool for high-throughput phenotyping of plant leaf traits due to its rapid, low-cost, multi-sensing, and non-destructive nature. However, collecting samples for model calibration can still be expensive, and models show poor transferability among different datasets. This study had three specific objectives: first, to assemble a large library of leaf hyperspectral data (n=2460) from maize and sorghum; second, to evaluate two machine-learning approaches to estimate nine leaf properties (chlorophyll, thickness, water content, nitrogen, phosphorus, potassium, calcium, magnesium, and sulfur); and third, to investigate the usefulness of this spectral library for predicting external datasets (n=445) including soybean and camelina using extra-weighted spiking. Internal cross-validation showed satisfactory performance of the spectral library to estimate all nine traits (mean R2=0.688), with partial least-squares regression outperforming deep neural network models. Models calibrated solely using the spectral library showed degraded performance on external datasets (mean R2=0.159 for camelina, 0.337 for soybean). Models improved significantly when a small portion of external samples (n=20) was added to the library via extra-weighted spiking (mean R2=0.574 for camelina, 0.536 for soybean). The leaf-level spectral library greatly benefits plant physiological and biochemical phenotyping, whilst extra-weight spiking improves model transferability and extends its utility.

A role for heritable transcriptomic variation in maize adaptation to temperate environments.

BACKGROUND: Transcription bridges genetic information and phenotypes. Here, we evaluated how changes in transcriptional regulation enable maize (Zea mays), a crop originally domesticated in the tropics, to adapt to temperate environments. RESULT: We generated 572 unique RNA-seq datasets from the roots of 340 maize genotypes. Genes involved in core processes such as cell division, chromosome organization and cytoskeleton organization showed lower heritability of gene expression, while genes involved in anti-oxidation activity exhibited higher expression heritability. An expression genome-wide association study (eGWAS) identified 19,602 expression quantitative trait loci (eQTLs) associated with the expression of 11,444 genes. A GWAS for alternative splicing identified 49,897 splicing QTLs (sQTLs) for 7614 genes. Genes harboring both cis-eQTLs and cis-sQTLs in linkage disequilibrium were disproportionately likely to encode transcription factors or were annotated as responding to one or more stresses. Independent component analysis of gene expression data identified loci regulating co-expression modules involved in oxidation reduction, response to water deprivation, plastid biogenesis, protein biogenesis, and plant-pathogen interaction. Several genes involved in cell proliferation, flower development, DNA replication, and gene silencing showed lower gene expression variation explained by genetic factors between temperate and tropical maize lines. A GWAS of 27 previously published phenotypes identified several candidate genes overlapping with genomic intervals showing signatures of selection during adaptation to temperate environments. CONCLUSION: Our results illustrate how maize transcriptional regulatory networks enable changes in transcriptional regulation to adapt to temperate regions.

Yield prediction through integration of genetic, environment, and management data through deep learning.

Accurate prediction of the phenotypic outcomes produced by different combinations of genotypes, environments, and management interventions remains a key goal in biology with direct applications to agriculture, research, and conservation. The past decades have seen an expansion of new methods applied toward this goal. Here we predict maize yield using deep neural networks, compare the efficacy of 2 model development methods, and contextualize model performance using conventional linear and machine learning models. We examine the usefulness of incorporating interactions between disparate data types. We find deep learning and best linear unbiased predictor (BLUP) models with interactions had the best overall performance. BLUP models achieved the lowest average error, but deep learning models performed more consistently with similar average error. Optimizing deep neural network submodules for each data type improved model performance relative to optimizing the whole model for all data types at once. Examining the effect of interactions in the best-performing model revealed that including interactions altered the model's sensitivity to weather and management features, including a reduction of the importance scores for timepoints expected to have a limited physiological basis for influencing yield-those at the extreme end of the season, nearly 200 days post planting. Based on these results, deep learning provides a promising avenue for the phenotypic prediction of complex traits in complex environments and a potential mechanism to better understand the influence of environmental and genetic factors.

A common resequencing-based genetic marker data set for global maize diversity.

Maize (Zea mays ssp. mays) populations exhibit vast ranges of genetic and phenotypic diversity. As sequencing costs have declined, an increasing number of projects have sought to measure genetic differences between and within maize populations using whole-genome resequencing strategies, identifying millions of segregating single-nucleotide polymorphisms (SNPs) and insertions/deletions (InDels). Unlike older genotyping strategies like microarrays and genotyping by sequencing, resequencing should, in principle, frequently identify and score common genetic variants. However, in practice, different projects frequently employ different analytical pipelines, often employ different reference genome assemblies and consistently filter for minor allele frequency within the study population. This constrains the potential to reuse and remix data on genetic diversity generated from different projects to address new biological questions in new ways. Here, we employ resequencing data from 1276 previously published maize samples and 239 newly resequenced maize samples to generate a single unified marker set of approximately 366 million segregating variants and approximately 46 million high-confidence variants scored across crop wild relatives, landraces as well as tropical and temperate lines from different breeding eras. We demonstrate that the new variant set provides increased power to identify known causal flowering-time genes using previously published trait data sets, as well as the potential to track changes in the frequency of functionally distinct alleles across the global distribution of modern maize.

Time series canopy phenotyping enables the identification of genetic variants controlling dynamic phenotypes in soybean.

Advances in plant phenotyping technologies are dramatically reducing the marginal costs of collecting multiple phenotypic measurements across several time points. Yet, most current approaches and best statistical practices implemented to link genetic and phenotypic variation in plants have been developed in an era of single-time-point data. Here, we used time-series phenotypic data collected with an unmanned aircraft system for a large panel of soybean (Glycine max (L.) Merr.) varieties to identify previously uncharacterized loci. Specifically, we focused on the dissection of canopy coverage (CC) variation from this rich data set. We also inferred the speed of canopy closure, an additional dimension of CC, from the time-series data, as it may represent an important trait for weed control. Genome-wide association studies (GWASs) identified 35 loci exhibiting dynamic associations with CC across developmental stages. The time-series data enabled the identification of 10 known flowering time and plant height quantitative trait loci (QTLs) detected in previous studies of adult plants and the identification of novel QTLs influencing CC. These novel QTLs were disproportionately likely to act earlier in development, which may explain why they were missed in previous single-time-point studies. Moreover, this time-series data set contributed to the high accuracy of the GWASs, which we evaluated by permutation tests, as evidenced by the repeated identification of loci across multiple time points. Two novel loci showed evidence of adaptive selection during domestication, with different genotypes/haplotypes favored in different geographic regions. In summary, the time-series data, with soybean CC as an example, improved the accuracy and statistical power to dissect the genetic basis of traits and offered a promising opportunity for crop breeding with quantitative growth curves.

The genome of Orychophragmus violaceus provides genomic insights into the evolution of Brassicaceae polyploidization and its distinct traits.

Orychophragmus violaceus, referred to as "eryuelan" (February orchid) in China, is an early-flowering ornamental plant. The high oil content and abundance of unsaturated fatty acids in O. violaceus seeds make it a potential high-quality oilseed crop. Here, we generated a whole-genome assembly for O. violaceus using Nanopore and Hi-C sequencing technologies. The assembled genome of O. violaceus was ∼1.3 Gb in size, with 12 pairs of chromosomes. Through investigation of ancestral genome evolution, we determined that the genome of O. violaceus experienced a tetraploidization event from a diploid progenitor with the translocated proto-Calepineae karyotype. Comparisons between the reconstructed subgenomes of O. violaceus identified indicators of subgenome dominance, indicating that subgenomes likely originated via allotetraploidy. O. violaceus was phylogenetically close to the Brassica genus, and tetraploidy in O. violaceus occurred approximately 8.57 million years ago, close in time to the whole-genome triplication of Brassica that likely arose via an intermediate tetraploid lineage. However, the tetraploidization in Orychophragmus was independent of the hexaploidization in Brassica, as evidenced by the results from detailed phylogenetic analyses and comparisons of the break and fusion points of ancestral genomic blocks. Moreover, identification of multi-copy genes regulating the production of high-quality oil highlighted the contributions of both tetraploidization and tandem duplication to functional innovation in O. violaceus. These findings provide novel insights into the polyploidization evolution of plant species and will promote both functional genomic studies and domestication/breeding efforts in O. violaceus.

Genome of Paspalum vaginatum and the role of trehalose mediated autophagy in increasing maize biomass.

A number of crop wild relatives can tolerate extreme stress to a degree outside the range observed in their domesticated relatives. However, it is unclear whether or how the molecular mechanisms employed by these species can be translated to domesticated crops. Paspalum (Paspalum vaginatum) is a self-incompatible and multiply stress-tolerant wild relative of maize and sorghum. Here, we describe the sequencing and pseudomolecule level assembly of a vegetatively propagated accession of P. vaginatum. Phylogenetic analysis based on 6,151 single-copy syntenic orthologues conserved in 6 related grass species places paspalum as an outgroup of the maize-sorghum clade. In parallel metabolic experiments, paspalum, but neither maize nor sorghum, exhibits a significant increase in trehalose when grown under nutrient-deficit conditions. Inducing trehalose accumulation in maize, imitating the metabolic phenotype of paspalum, results in autophagy dependent increases in biomass accumulation.