Search

In plant breeding and genetics, predictive models traditionally rely on compact representations of high-dimensional data, often using methods like Principal Component Analysis (PCA) and, more recently, Autoencoders (AE). However, these methods do not separate genotype-specific and environment-specific features, limiting their ability to accurately predict traits influenced by both genetic and environmental factors. We hypothesize that disentangling these representations into genotype-specific and environment-specific components can enhance predictive models. To test this, we developed a compositional autoencoder (CAE) that decomposes high-dimensional data into distinct genotype-specific and environment-specific latent features. Our CAE framework employed a hierarchical architecture within an autoencoder to effectively separate these entangled latent features. Applied to a maize diversity panel dataset, the CAE demonstrated superior modeling of environmental influences and out-performs PCA (principal component analysis), PLSR (Partial Least square regression) and vanilla autoencoders by 7 times for ‘Days to Pollen’ trait and 10 times improved predictive performance for ‘Yield’. By disentangling latent features, the CAE provided a powerful tool for precision breeding and genetic research. This work has significantly enhanced trait prediction models, advancing agricultural and biological sciences.

Abstract Sunflower (Helianthus annuus L.) plays an essential role in meeting the demand for edible oil worldwide. The yield of sunflower seeds encompasses several component traits, including the disc diameter. Over three consecutive years, 2019, 2020, and 2022, we assessed phenotypic variation in disc diameter across a diverse set of sunflower accessions (N = 342) in replicated field trials. Upon aggregating the phenotypic data from multiple years, we estimated the broad sense heritability (H2) of the disc diameter trait to be 0.88. A subset of N = 274 accessions was genotyped by using the tunable genotyping‐by‐sequencing (tGBS) method, resulting in 226,779 high‐quality SNPs. Using these SNPs and the disc diameter phenotype, we conducted a genome‐wide association study (GWAS) employing two statistical approaches: the mixed linear model (MLM) and the fixed and random model circulating probability unification (farmCPU). The MLM and farmCPU GWAS approaches identified 106 and 8 significant SNPs located close to 53 and 21 genes, respectively. The MLM analysis identified two significant peaks: a prominent signal on chromosome 10 and a relatively weaker signal on chromosome 16, both of which were also detected by farmCPU. The genetic loci associated with disc diameter, as well as the related candidate genes, present promising avenues for further functional validation and serve as a basis for sunflower oil yield improvement.

Abstract 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.

Abstract 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.

Abstract 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.

Paspalum vaginatum is a stress tolerant wild relative of maize and sorghum. Here, the authors assemble its genome at pseudomolecule level and reveal the role of trehalose mediated autophagy in increasing maize biomass productivity under nutrient-deficit conditions.

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.

Abstract Background Access to biologically available nitrogen is a key constraint on plant growth in both natural and agricultural settings. Variation in tolerance to nitrogen deficit stress and productivity in nitrogen limited conditions exists both within and between plant species. However, our understanding of changes in different phenotypes under long term low nitrogen stress and their impact on important agronomic traits, such as yield, is still limited. Results Here we quantified variation in the metabolic, physiological, and morphological responses of a sorghum association panel assembled to represent global genetic diversity to long term, nitrogen deficit stress and the relationship of these responses to grain yield under both conditions. Grain yield exhibits substantial genotype by environment interaction while many other morphological and physiological traits exhibited consistent responses to nitrogen stress across the population. Large scale nontargeted metabolic profiling for a subset of lines in both conditions identified a range of metabolic responses to long term nitrogen deficit stress. Several metabolites were associated with yield under high and low nitrogen conditions. Conclusion Our results highlight that grain yield in sorghum, unlike many morpho-physiological traits, exhibits substantial variability of genotype specific responses to long term low severity nitrogen deficit stress. Metabolic response to long term nitrogen stress shown higher proportion of variability explained by genotype specific responses than did morpho-pysiological traits and several metabolites were correlated with yield. This suggest, that it might be possible to build predictive models using metabolite abundance to estimate which sorghum genotypes will exhibit greater or lesser decreases in yield in response to nitrogen deficit, however further research needs to be done to evaluate such model.

Diet affects the human gut microbiome, but studies linking crop genetics to seed traits that influence the human gut microbiome are lacking. Here, the authors develop an in vitro microbiome screening method and reveal the association between sorghum genes regulating condensed tannin biosynthesis and human gut microbiome.

The small size (≤15-nt) of micorexons poses difficulties for genome annotation and identification using standard RNA sequence mapping approaches. Here, the authors develop computational pipelines to discover and predict microexons in plants and reveal diverse evolutionary trajectories via genomewide microexon modeling.

Abstract Domesticated ~10,000 years ago in northern China, Proso millet (Panicum miliaceum L.) is a climate‐resilient and human health‐promoting cereal crop. The genome size of this self‐pollinated allotetraploid is 923 Mb. Proso millet seeds are an important part of the human diet in many countries. In the USA, its use is restricted to the birdseed and pet food market. Proso millet is witnessing gradual demand in the global human health and wellness food market owing to its health‐promoting properties such as low glycemic index and gluten‐free. The breeding efforts for developing improved proso millet cultivars are hindered by the dearth of genomic resources available to researchers. The publication of the reference genome and availability of cost‐effective NGS methodologies could lead to the identification of high‐quality genetic variants, which can be incorporated into breeding pipelines. Here, we report the identification of single‐nucleotide polymorphisms (SNPs) by low‐pass (1×) genome sequencing of 85 diverse proso millet accessions from 23 different countries. The 2 × 150 bp Illumina paired‐end reads generated after sequencing were aligned to the proso millet reference genome. The resulting sequence alignment information was used to call SNPs. We obtained 972,863 bi‐allelic SNPs after quality filtering of the raw SNPs. These SNPs were used to assess the population structure and phylogenetic relationships among the accessions. Most of the accessions were found to be highly inbred with heterozygosity ranging between .05 and .20. Principal component analysis (PCA) showed that PC1 (principal component) and PC2 explained 19% of the variability in the population. PCA also clustered all the genotypes into three groups. A neighbor‐joining tree clustered the genotypes into four distinct groups exhibiting diverse representation within the population. The SNPs identified in our study could be used for molecular breeding and genetics research (e.g., genetic and association mapping, and population genetics) in proso millet after proper validation.

The effects of fiber, complex carbohydrates, lipids, and small molecules from food matrices on the human gut microbiome have been increasingly studied. Much less is known about how dietary protein can influence the composition and function of the gut microbial community. Here, we used near-isogenic maize lines of conventional popcorn and quality-protein popcorn (QPP) to study the effects of the opaque-2 mutation and associated quality-protein modifiers on the human gut microbiome. Opaque-2 blocks the synthesis of major maize seed proteins (α-zeins), resulting in a compensatory synthesis of new seed proteins that are nutritionally beneficial with substantially higher levels of the essential amino acids lysine and tryptophan. We show that QPP lines stimulate greater amounts of butyrate production by human gut microbiomes in in vitro fermentation of popped and digested corn from parental and QPP hybrids. In human gut microbiomes derived from diverse individuals, bacterial taxa belonging to the butyrate-producing family Lachnospiraceae, including the genera Coprococcus and Roseburia were consistently increased when fermenting QPP vs. parental popcorn lines. We conducted molecular complementation to further demonstrate that lysine-enriched seed protein can stimulate growth and butyrate production by microbes through distinct pathways. Our data show that organisms such as Coprococcus can utilize lysine and that other gut microbes, such as Roseburia spp., instead, utilize fructoselysine produced during thermal processing (popping) of popcorn. Thus, the combination of seed composition in QPP and interaction of protein adducts with carbohydrates during thermal processing can stimulate the growth of health-promoting, butyrate-producing organisms in the human gut microbiome through multiple pathways.

Abstract Objectives The purpose of this data set is to capture the complete diurnal (i.e., daily) transcriptome of fully expanded third leaves from the C4 panacoid grasses sorghum (Sorghum bicolor), maize (Zea mays), and foxtail millet (Setaria italica) with RNA-seq transcriptome profiling. These data are the cornerstone of a larger project that examined the conservation and divergence of gene expression networks within these crop plants. This data set focuses on temporal changes in gene expression to identify the network architecture responsible for daily regulation of plant growth and metabolic activities. The power of this data set is fine temporal resolution combined with continuous sampling over multiple days. Data description The data set is 72 individual RNA-seq samples representing 24 time course samples each for sorghum, maize, and foxtail millet plants cultivated in a growth chamber under equal intervals of light and darkness. The 24 samples are separated by 3-h intervals so that the data set is a fine scale 72-h analysis of gene expression in the leaves of each plant type. FASTQ files from Illumina sequencing are available at the National Center for Biotechnology Information Sequence Read Archive.

Abstract Background The circadian clock drives endogenous 24-h rhythms that allow organisms to adapt and prepare for predictable and repeated changes in their environment throughout the day-night (diurnal) cycle. Many components of the circadian clock in Arabidopsis thaliana have been functionally characterized, but comparatively little is known about circadian clocks in grass species including major crops like maize and sorghum. Results Comparative research based on protein homology and diurnal gene expression patterns suggests the function of some predicted clock components in grasses is conserved with their Arabidopsis counterparts, while others have diverged in function. Our analysis of diurnal gene expression in three panicoid grasses sorghum, maize, and foxtail millet revealed conserved and divergent evolution of expression for core circadian clock genes and for the overall transcriptome. We find that several classes of core circadian clock genes in these grasses differ in copy number compared to Arabidopsis, but mostly exhibit conservation of both protein sequence and diurnal expression pattern with the notable exception of maize paralogous genes. We predict conserved cis-regulatory motifs shared between maize, sorghum, and foxtail millet through identification of diurnal co-expression clusters for a subset of 27,196 orthologous syntenic genes. In this analysis, a Cochran–Mantel–Haenszel based method to control for background variation identified significant enrichment for both expected and novel 6–8 nucleotide motifs in the promoter regions of genes with shared diurnal regulation predicted to function in common physiological activities. Conclusions This study illustrates the divergence and conservation of circadian clocks and diurnal regulatory networks across syntenic orthologous genes in panacoid grass species. Further, conserved local regulatory sequences contribute to the architecture of these diurnal regulatory networks that produce conserved patterns of diurnal gene expression.

Abstract Objectives Advanced tools and resources are needed to efficiently and sustainably produce food for an increasing world population in the context of variable environmental conditions. The maize genomes to fields (G2F) initiative is a multi-institutional initiative effort that seeks to approach this challenge by developing a flexible and distributed infrastructure addressing emerging problems. G2F has generated large-scale phenotypic, genotypic, and environmental datasets using publicly available inbred lines and hybrids evaluated through a network of collaborators that are part of the G2F’s genotype-by-environment (G × E) project. This report covers the public release of datasets for 2014–2017. Data description Datasets include inbred genotypic information; phenotypic, climatic, and soil measurements and metadata information for each testing location across years. For a subset of inbreds in 2014 and 2015, yield component phenotypes were quantified by image analysis. Data released are accompanied by README descriptions. For genotypic and phenotypic data, both raw data and a version without outliers are reported. For climatic data, a version calibrated to the nearest airport weather station and a version without outliers are reported. The 2014 and 2015 datasets are updated versions from the previously released files [1] while 2016 and 2017 datasets are newly available to the public.

Selection for yield at high planting density has reshaped the leaf canopy of maize, improving photosynthetic productivity in high density settings. Further optimization of canopy architecture may be possible. However, measuring leaf angles, the widely studied component trait of leaf canopy architecture, by hand is a labor and time intensive process. Here, we use multiple, calibrated, 2D images to reconstruct the 3D geometry of individual sorghum plants using a voxel carving based algorithm. Automatic skeletonization and segmentation of these 3D geometries enable quantification of the angle of each leaf for each plant. The resulting measurements are both heritable and correlated with manually collected leaf angles. This automated and scaleable reconstruction approach was employed to measure leaf-by-leaf angles for a population of 366 sorghum plants at multiple time points, resulting in 971 successful reconstructions and 3,376 leaf angle measurements from individual leaves. A genome wide association study conducted using aggregated leaf angle data identified a known large effect leaf angle gene, several previously identified leaf angle QTL from a sorghum NAM population, and novel signals. Genome wide association studies conducted separately for three individual sorghum leaves identified a number of the same signals, a previously unreported signal shared across multiple leaves, and signals near the sorghum orthologs of two maize genes known to influence leaf angle. Automated measurement of individual leaves and mapping variants associated with leaf angle reduce the barriers to engineering ideal canopy architectures in sorghum and other grain crops.

Genome-wide DNA polymorphism analysis and molecular marker development are important for forward genetics research and DNA marker-assisted breeding. As an ideal model system for Panicoideae grasses and an important minor crop in East Asia, foxtail millet (Setaria italica) has a high-quality reference genome as well as large mutant libraries based on the “Yugu1” variety. However, there is still a lack of genetic and mutation mapping tools available for forward genetics research on S. italica. Here, we screened another S. italica genotype, “SSR41”, which is morphologically similar to, and readily cross-pollinates with, “Yugu1”. High-throughput resequencing of “SSR41” identified 1,102,064 reliable single nucleotide polymorphisms (SNPs) and 196,782 insertions/deletions (InDels) between the two genotypes, indicating that these two genotypes have high genetic diversity. Of the 8,361 high-quality InDels longer than 20 bp that were developed as molecular markers, 180 were validated with 91.5% accuracy. We used “SSR41” and these developed molecular markers to map the white leaf sheath gene SiWLS1. Further analyses showed that SiWLS1 encodes a chloroplast-localized protein that is involved in the regulation of chloroplast development in bundle sheath cells in the leaf sheath in S. italica and is related to sensitivity to heavy metals. Our study provides the methodology and an important resource for forward genetics research on Setaria.

Abstract As a step towards trait mapping in the halophyte seashore paspalum (Paspalum vaginatum Sw.), we developed an F1 mapping population from a cross between two genetically diverse and heterozygous accessions, 509022 and HI33. Progeny were genotyped using a genotyping-by-sequencing (GBS) approach and sequence reads were analyzed for single nucleotide polymorphisms (SNPs) using the UGbS-Flex pipeline. More markers were identified that segregated in the maternal parent (HA maps) compared to the paternal parent (AH maps), suggesting that 509022 had overall higher levels of heterozygosity than HI33. We also generated maps that consisted of markers that were heterozygous in both parents (HH maps). The AH, HA and HH maps each comprised more than 1000 markers. Markers formed 10 linkage groups, corresponding to the ten seashore paspalum chromosomes. Comparative analyses showed that each seashore paspalum chromosome was syntenic to and highly colinear with a single sorghum chromosome. Four inversions were identified, two of which were sorghum-specific while the other two were likely specific to seashore paspalum. These high-density maps are the first available genetic maps for seashore paspalum. The maps will provide a valuable tool for plant breeders and others in the Paspalum community to identify traits of interest, including salt tolerance.

Abstract Background Hyperspectral reflectance data in the visible, near infrared and shortwave infrared range (VIS–NIR–SWIR, 400–2500 nm) are commonly used to nondestructively measure plant leaf properties. We investigated the usefulness of VIS–NIR–SWIR as a high-throughput tool to measure six leaf properties of maize plants including chlorophyll content (CHL), leaf water content (LWC), specific leaf area (SLA), nitrogen (N), phosphorus (P), and potassium (K). This assessment was performed using the lines of the maize diversity panel. Data were collected from plants grown in greenhouse condition, as well as in the field under two nitrogen application regimes. Leaf-level hyperspectral data were collected with a VIS–NIR–SWIR spectroradiometer at tasseling. Two multivariate modeling approaches, partial least squares regression (PLSR) and support vector regression (SVR), were employed to estimate the leaf properties from hyperspectral data. Several common vegetation indices (VIs: GNDVI, RENDVI, and NDWI), which were calculated from hyperspectral data, were also assessed to estimate these leaf properties. Results Some VIs were able to estimate CHL and N (R2 > 0.68), but failed to estimate the other four leaf properties. Models developed with PLSR and SVR exhibited comparable performance to each other, and provided improved accuracy relative to VI models. CHL were estimated most successfully, with R2 (coefficient of determination) > 0.94 and ratio of performance to deviation (RPD) > 4.0. N was also predicted satisfactorily (R2 > 0.85 and RPD > 2.6). LWC, SLA and K were predicted moderately well, with R2 ranging from 0.54 to 0.70 and RPD from 1.5 to 1.8. The lowest prediction accuracy was for P, with R2

Broomcorn millet is one of the earliest domesticated plants and has the highest water use efficiency among cereals. Here, the authors report its genome assembly and annotation, which provides a valuable resource for breeders and paves the way for studying plant drought tolerance and C4 photosynthesis.

The timing of flowering plays a critical role in determining the productivity of agricultural crops. If the crops flower too early, the crop would mature before the end of the growing season, losing the opportunity to capture and use large amounts of light energy. If the crops flower too late, the crop may be killed by the change of seasons before it is ready to harvest. Maize flowering is one of the most important periods where even small amounts of stress can significantly alter yield. In this work, we developed and compared two methods for automatic tassel detection based on the imagery collected from an unmanned aerial vehicle, using deep learning models. The first approach was a customized framework for tassel detection based on convolutional neural network (TD-CNN). The other method was a state-of-the-art object detection technique of the faster region-based CNN (Faster R-CNN), serving as baseline detection accuracy. The evaluation criteria for tassel detection were customized to correctly reflect the needs of tassel detection in an agricultural setting. Although detecting thin tassels in the aerial imagery is challenging, our results showed promising accuracy: the TD-CNN had an F1 score of 95.9% and the Faster R-CNN had 97.9% F1 score. More CNN-based model structures can be investigated in the future for improved accuracy, speed, and generalizability on aerial-based tassel detection.

Genomic prediction provides an efficient alternative to conventional phenotypic selection for developing improved cultivars with desirable characteristics. New and improved methods to genomic prediction are continually being developed that attempt to deal with the integration of data types beyond genomic information. Modern automated weather systems offer the opportunity to capture continuous data on a range of environmental parameters at specific field locations. In principle, this information could characterize training and target environments and enhance predictive ability by incorporating weather characteristics as part of the genotype-by-environment (G×E) interaction component in prediction models. We assessed the usefulness of including weather data variables in genomic prediction models using a naïve environmental kinship model across 30 environments comprising the Genomes to Fields (G2F) initiative in 2014 and 2015. Specifically four different prediction scenarios were evaluated (i) tested genotypes in observed environments; (ii) untested genotypes in observed environments; (iii) tested genotypes in unobserved environments; and (iv) untested genotypes in unobserved environments. A set of 1,481 unique hybrids were evaluated for grain yield. Evaluations were conducted using five different models including main effect of environments; general combining ability (GCA) effects of the maternal and paternal parents modeled using the genomic relationship matrix; specific combining ability (SCA) effects between maternal and paternal parents; interactions between genetic (GCA and SCA) effects and environmental effects; and finally interactions between the genetics effects and environmental covariates. Incorporation of the genotype-by-environment interaction term improved predictive ability across all scenarios. However, predictive ability was not improved through inclusion of naive environmental covariates in G×E models. More research should be conducted to link the observed weather conditions with important physiological aspects in plant development to improve predictive ability through the inclusion of weather data.

Abstract Tandem duplication gives rise to copy number variation and subsequent functional novelty among genes as well as diversity between individuals in a species. Functional novelty can result from either divergence in coding sequence or divergence in patterns of gene transcriptional regulation. Here, we investigate conservation and divergence of both gene sequence and gene regulation between the copies of the α‐zein gene family in maize inbreds B73 and W22. We used RNA‐seq data generated from developing, self‐pollinated kernels at three developmental stages timed to coincide with early and peak zein expression. The reference genome annotations for B73 and W22 were modified to ensure accurate inclusion of their respective α‐zein gene models to accurately assess copy‐specific expression. Expression analysis indicated that although the total expression of α‐zeins is higher in W22, the pattern of expression in both lines is conserved. Additional analysis of publicly available RNA‐seq data from a diverse population of maize inbreds also demonstrates variation in absolute expression, but conservation of expression patterns across a wide range of maize genotypes and α‐zein haplotypes.

Abstract Leaf number and leaf emergence rate are phenotypes of interest to plant breeders, plant geneticists, and crop modelers. Counting the extant leaves of an individual plant is straightforward even for an untrained individual, but manually tracking changes in leaf numbers for hundreds of individuals across multiple time points is logistically challenging. This study generated a dataset including over 150,000 maize and sorghum images for leaf counting projects. A subset of 17,783 images also includes annotations of the positions of individual leaf tips. With these annotated images, we evaluate two deep learning‐based approaches for automated leaf counting: the first based on counting‐by‐regression from whole image analysis and a second based on counting‐by‐detection. Both approaches can achieve root of mean square error (RMSE) smaller than one leaf, only moderately inferior to the RMSE between human annotators of between 0.57 and 0.73 leaves. The counting‐by‐regression approach based on convolutional neural networks (CNNs) exhibited lower accuracy and increased bias for plants with extreme leaf numbers which are underrepresented in this dataset. The counting‐by‐detection approach based on Faster R‐CNNs (region based convolutional neural networks) object detection models achieve near human performance for plants where all leaf tips are visible. The annotated image data and model performance metrics generated as part of this study provide large scale resources for the comparison and improvement of algorithms for leaf counting from image data in grain crops.

Abstract Changes in canopy architecture traits have been shown to contribute to yield increases. Optimizing both light interception and light interception efficiency of agricultural crop canopies will be essential to meeting the growing food needs. Canopy architecture is inherently three‐dimensional (3D), but many approaches to measuring canopy architecture component traits treat the canopy as a two‐dimensional (2D) structure to make large scale measurement, selective breeding, and gene identification logistically feasible. We develop a high throughput voxel carving strategy to reconstruct 3D representations of sorghum from a small number of RGB photos. Our approach builds on the voxel carving algorithm to allow for fully automatic reconstruction of hundreds of plants. It was employed to generate 3D reconstructions of individual plants within a sorghum association population at the late vegetative stage of development. Light interception parameters estimated from these reconstructions enabled the identification of known and previously unreported loci controlling light interception efficiency in sorghum. The approach is generalizable and scalable, and it enables 3D reconstructions from existing plant high throughput phenotyping datasets. We also propose a set of best practices to increase 3D reconstructions’ accuracy.

Premise Maize yields have significantly increased over the past half‐century owing to advances in breeding and agronomic practices. Plants have been grown in increasingly higher densities due to changes in plant architecture resulting in plants with more upright leaves, which allows more efficient light interception for photosynthesis. Natural variation for leaf angle has been identified in maize and sorghum using multiple mapping populations. However, conventional phenotyping techniques for leaf angle are low throughput and labor intensive, and therefore hinder a mechanistic understanding of how the leaf angle of individual leaves changes over time in response to the environment. Methods High‐throughput time series image data from water‐deprived maize (Zea mays subsp. mays) and sorghum (Sorghum bicolor) were obtained using battery‐powered time‐lapse cameras. A MATLAB‐based image processing framework, Leaf Angle eXtractor (LAX), was developed to extract and quantify leaf angles from images of maize and sorghum plants under drought conditions. Results Leaf angle measurements showed differences in leaf responses to drought in maize and sorghum. Tracking leaf angle changes at intervals as short as one minute enabled distinguishing leaves that showed signs of wilting under water deprivation from other leaves on the same plant that did not show wilting during the same time period. Discussion Automating leaf angle measurements using LAX makes it feasible to perform large‐scale experiments to evaluate, understand, and exploit the spatial and temporal variations in plant response to water limitations.

Abstract Advances in genome sequencing and annotation have eased the difficulty of identifying new gene sequences. Predicting the functions of these newly identified genes remains challenging. Genes descended from a common ancestral sequence are likely to have common functions. As a result, homology is widely used for gene function prediction. This means functional annotation errors also propagate from one species to another. Several approaches based on machine learning classification algorithms were evaluated for their ability to accurately predict gene function from non‐homology gene features. Among the eight supervised classification algorithms evaluated, random‐forest‐based prediction consistently provided the most accurate gene function prediction. Non‐homology‐based functional annotation provides complementary strengths to homology‐based annotation, with higher average performance in Biological Process GO terms, the domain where homology‐based functional annotation performs the worst, and weaker performance in Molecular Function GO terms, the domain where the accuracy of homology‐based functional annotation is highest. GO prediction models trained with homology‐based annotations were able to successfully predict annotations from a manually curated “gold standard” GO annotation set. Non‐homology‐based functional annotation based on machine learning may ultimately prove useful both as a method to assign predicted functions to orphan genes which lack functionally characterized homologs, and to identify and correct functional annotation errors which were propagated through homology‐based functional annotations.

Abstract The number of plant species with genomic and transcriptomic data has been increasing rapidly. The grasses—Poaceae—have been well represented among species with published reference genomes. However, as a result the genomes of wild grasses are less frequently targeted by sequencing efforts. Sequence data from wild relatives of crop species in the grasses can aid the study of domestication, gene discovery for breeding and crop improvement, and improve our understanding of the evolution of C4 photosynthesis. Here, we used long‐read sequencing technology to characterize the transcriptomes of three C3 panicoid grass species: Dichanthelium oligosanthes, Chasmanthium laxum, and Hymenachne amplexicaulis. Based on alignments to the sorghum genome, we estimate that assembled consensus transcripts from each species capture between 54.2% and 65.7% of the conserved syntenic gene space in grasses. Genes co‐opted into C4 were also well represented in this dataset, despite concerns that because these genes might play roles unrelated to photosynthesis in the target species, they would be expressed at low levels and missed by transcript‐based sequencing. A combined analysis using syntenic orthologous genes from grasses with published reference genomes and consensus long‐read sequences from these wild species was consistent with previously published phylogenies. It is hoped that these data, targeting underrepresented classes of species within the PACMAD grasses—wild species and species utilizing C3 photosynthesis—will aid in future studies of domestication and C4 evolution by decreasing the evolutionary distance between C4 and C3 species within this clade, enabling more accurate comparisons associated with evolution of the C4 pathway.

High-throughput phenotyping system has become more and more popular in plant science research. The data analysis for such a system typically involves two steps: plant feature extraction through image processing and statistical analysis for the extracted features. The current approach is to perform those two steps on different platforms. We develop the package “implant” in R for both robust feature extraction and functional data analysis. For image processing, the “implant” package provides methods including thresholding, hidden Markov random field model, and morphological operations. For statistical analysis, this package can produce nonparametric curve fitting with its confidence region for plant growth. A functional ANOVA model to test for the treatment and genotype effects on the plant growth dynamics is also provided.

This study describes the evaluation of a range of approaches to semantic segmentation of hyperspectral images of sorghum plants, classifying each pixel as either nonplant or belonging to one of the three organ types (leaf, stalk, panicle). While many current methods for segmentation focus on separating plant pixels from background, organ-specific segmentation makes it feasible to measure a wider range of plant properties. Manually scored training data for a set of hyperspectral images collected from a sorghum association population was used to train and evaluate a set of supervised classification models. Many algorithms show acceptable accuracy for this classification task. Algorithms trained on sorghum data are able to accurately classify maize leaves and stalks, but fail to accurately classify maize reproductive organs which are not directly equivalent to sorghum panicles. Trait measurements extracted from semantic segmentation of sorghum organs can be used to identify both genes known to be controlling variation in a previously measured phenotypes (e.g., panicle size and plant height) as well as identify signals for genes controlling traits not previously quantified in this population (e.g., stalk/leaf ratio). Organ level semantic segmentation provides opportunities to identify genes controlling variation in a wide range of morphological phenotypes in sorghum, maize, and other related grain crops.

Genomic selection (GS) is an emerging methodology that helps select superior lines among experimental cultivars in plant breeding programs. It offers the opportunity to increase the productivity of cultivars by delivering increased genetic gains and reducing the breeding cycles. This methodology requires inexpensive and sufficiently dense marker information to be successful, and with whole genome sequencing, it has become an important tool in many crops. The recent assembly of the pearl millet genome has made it possible to employ GS models to improve the selection procedure in pearl millet breeding programs. Here, three GS models were implemented and compared using grain yield and dense molecular marker information of pearl millet obtained from two different genotyping platforms (C [conventional GBS RAD-seq] and T [tunable GBS tGBS]). The models were evaluated using three different cross-validation (CV) schemes mimicking real situations that breeders face in breeding programs: CV2 resembles an incomplete field trial, CV1 predicts the performance of untested hybrids, and CV0 predicts the performance of hybrids in unobserved environments. We found that (i) adding phenotypic information of parental inbreds to the calibration sets improved predictive ability, (ii) accounting for genotype-by-environment interaction also increased the performance of the models, and (iii) superior strategies should consider the use of the molecular markers derived from the T platform (tGBS).

Abstract Background Short read DNA sequencing technologies have revolutionized genome assembly by providing high accuracy and throughput data at low cost. But it remains challenging to assemble short read data, particularly for large, complex and polyploid genomes. The linked read strategy has the potential to enhance the value of short reads for genome assembly because all reads originating from a single long molecule of DNA share a common barcode. However, the majority of studies to date that have employed linked reads were focused on human haplotype phasing and genome assembly. Results Here we describe a de novo maize B73 genome assembly generated via linked read technology which contains ~ 172,000 scaffolds with an N50 of 89 kb that cover 50% of the genome. Based on comparisons to the B73 reference genome, 91% of linked read contigs are accurately assembled. Because it was possible to identify errors with > 76% accuracy using machine learning, it may be possible to identify and potentially correct systematic errors. Complex polyploids represent one of the last grand challenges in genome assembly. Linked read technology was able to successfully resolve the two subgenomes of the recent allopolyploid, proso millet (Panicum miliaceum). Our assembly covers ~ 83% of the 1 Gb genome and consists of 30,819 scaffolds with an N50 of 912 kb. Conclusions Our analysis provides a framework for future de novo genome assemblies using linked reads, and we suggest computational strategies that if implemented have the potential to further improve linked read assemblies, particularly for repetitive genomes.

Pearl millet is a non-model grain and fodder crop adapted to extremely hot and dry environments globally. In India, a great deal of public and private sectors’ investment has focused on developing pearl millet single cross hybrids based on the cytoplasmic-genetic male sterility (CMS) system, while in Africa most pearl millet production relies on open pollinated varieties. Pearl millet lines were phenotyped for both the inbred parents and hybrids stage. Many breeding efforts focus on phenotypic selection of inbred parents to generate improved parental lines and hybrids. This study evaluated two genotyping techniques and four genomic selection schemes in pearl millet. Despite the fact that 6× more sequencing data were generated per sample for RAD-seq than for tGBS, tGBS yielded more than 2× as many informative SNPs (defined as those having MAF > 0.05) than RAD-seq. A genomic prediction scheme utilizing only data from hybrids generated prediction accuracies (median) ranging from 0.73-0.74 (1000-grain weight), 0.87-0.89 (days to flowering time), 0.48-0.51 (grain yield) and 0.72-0.73 (plant height). For traits with little to no heterosis, hybrid only and hybrid/inbred prediction schemes performed almost equivalently. For traits with significant mid-parent heterosis, the direct inclusion of phenotypic data from inbred lines significantly (P < 0.05) reduced prediction accuracy when all lines were analyzed together. However, when inbreds and hybrid trait values were both scored relative to the mean trait values for the respective populations, the inclusion of inbred phenotypic datasets moderately improved genomic predictions of the hybrid genomic estimated breeding values. Here we show that modern approaches to genotyping by sequencing can enable genomic selection in pearl millet. While historical pearl millet breeding records include a wealth of phenotypic data from inbred lines, we demonstrate that the naive incorporation of this data into a hybrid breeding program can reduce prediction accuracy, while controlling for the effects of heterosis per se allowed inbred genotype and trait data to improve the accuracy of genomic estimated breeding values for pearl millet hybrids.

Abstract Objectives Crop improvement relies on analysis of phenotypic, genotypic, and environmental data. Given large, well-integrated, multi-year datasets, diverse queries can be made: Which lines perform best in hot, dry environments? Which alleles of specific genes are required for optimal performance in each environment? Such datasets also can be leveraged to predict cultivar performance, even in uncharacterized environments. The maize Genomes to Fields (G2F) Initiative is a multi-institutional organization of scientists working to generate and analyze such datasets from existing, publicly available inbred lines and hybrids. G2F’s genotype by environment project has released 2014 and 2015 datasets to the public, with 2016 and 2017 collected and soon to be made available. Data description Datasets include DNA sequences; traditional phenotype descriptions, as well as detailed ear, cob, and kernel phenotypes quantified by image analysis; weather station measurements; and soil characterizations by site. Data are released as comma separated value spreadsheets accompanied by extensive README text descriptions. For genotypic and phenotypic data, both raw data and a version with outliers removed are reported. For weather data, two versions are reported: a full dataset calibrated against nearby National Weather Service sites and a second calibrated set with outliers and apparent artifacts removed.

Abstract Background Maize was originally domesticated in a tropical environment but is now widely cultivated at temperate latitudes. Temperate and tropical maize populations have diverged both genotypically and phenotypically. Tropical maize lines grown in temperate environments usually exhibit delayed flowering, pollination, and seed set, which reduces their grain yield relative to temperate adapted maize lines. One potential mechanism by which temperate maize may have adapted to a new environment is novel transposable element insertions, which can influence gene regulation. Recent advances in sequencing technology have made it possible to study variation in transposon content and insertion location in large sets of maize lines. Results In total, 274,408 non-redundant TEs (NRTEs) were identified using resequencing data generated from 83 maize inbred lines. The locations of DNA TEs and copia-superfamily retrotransposons showed significant positive correlations with gene density and genetic recombination rates, whereas gypsy-superfamily retrotransposons showed a negative correlation with these two parameters. Compared to tropical maize, temperate maize had fewer unique NRTEs but higher insertion frequency, lower background recombination rates, and higher linkage disequilibrium, with more NRTEs close to flowering and stress-related genes in the genome. Association mapping demonstrated that the presence/absence of 48 NRTEs was associated with flowering time and that expression of neighboring genes differed between haplotypes where a NRTE was present or absent. Conclusions This study suggests that NRTEs may have played an important role in creating the variation in gene regulation that enabled the rapid adaptation of maize to diverse environments.

Recent advances in automated plant phenotyping have enabled the collection of time series measurements from the same plants of a wide range of traits at different developmental time scales. The availability of time series phenotypic datasets has increased interest in statistical approaches for comparing patterns of change among different plant genotypes and different treatment conditions. Two widely used methods of modeling growth with time are pointwise analysis of variance (ANOVA) and parametric sigmoidal curve fitting. Pointwise ANOVA yields discontinuous growth curves, which do not reflect the true dynamics of growth patterns in plants. In contrast, fitting a parametric model to a time series of observations does capture the trend of growth; however, these models require assumptions regarding the true pattern of plant growth. Depending on the species, treatment regime, and subset of the plant life cycle sampled, these assumptions will not always hold true. We have developed a different approach—functional ANOVA—which yields continuous growth curves without requiring assumptions regarding patterns of plant growth. We compared and validated this approach using data from an experiment measuring the growth of two maize ( L. ssp. ) genotypes under two water availability treatments during a 21-d period. Functional ANOVA enables a nonparametric estimation of the dynamics of changes in plant traits with time without assumptions regarding curve shape. In addition to estimating smooth curves of trait values with time, functional ANOVA also estimates the derivatives of these curves, e.g., growth rates, simultaneously. Using two different subsampling strategies, we demonstrate that this functional ANOVA method enables the comparison of growth curves among plants phenotyped on non-overlapping days with little reduction in estimation accuracy. This means that functional ANOVA based approaches can allow larger numbers of samples and biological replicates to be scored in a single experiment given fixed amounts of phenotyping infrastructure and personnel.

Image-based high-throughput plant phenotyping in greenhouse has the potential to relieve the bottleneck currently presented by phenotypic scoring which limits the throughput of gene discovery and crop improvement efforts. Numerous studies have employed automated RGB imaging to characterize biomass and growth of agronomically important crops. The objective of this study was to investigate the utility of hyperspectral imaging for quantifying chemical properties of maize and soybean plants in vivo. These properties included leaf water content, as well as concentrations of macronutrients nitrogen (N), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca), and sulfur (S), and micronutrients sodium (Na), iron (Fe), manganese (Mn), boron (B), copper (Cu), and zinc (Zn). Hyperspectral images were collected from 60 maize and 60 soybean plants, each subjected to varying levels of either water deficit or nutrient limitation stress with the goal of creating a wide range of variation in the chemical properties of plant leaves. Plants were imaged on an automated conveyor belt system using a hyperspectral imager with a spectral range from 550 to 1,700 nm. Images were processed to extract reflectance spectrum from each plant and partial least squares regression models were developed to correlate spectral data with chemical data. Among all the chemical properties investigated, water content was predicted with the highest accuracy [R2 = 0.93 and RPD (Ratio of Performance to Deviation) = 3.8]. All macronutrients were also quantified satisfactorily (R2 from 0.69 to 0.92, RPD from 1.62 to 3.62), with N predicted best followed by P, K, and S. The micronutrients group showed lower prediction accuracy (R2 from 0.19 to 0.86, RPD from 1.09 to 2.69) than the macronutrient groups. Cu and Zn were best predicted, followed by Fe and Mn. Na and B were the only two properties that hyperspectral imaging was not able to quantify satisfactorily (R2 < 0.3 and RPD < 1.2). This study suggested the potential usefulness of hyperspectral imaging as a high-throughput phenotyping technology for plant chemical traits. Future research is needed to test the method more thoroughly by designing experiments to vary plant nutrients individually and cover more plant species, genotypes, and growth stages.

Identifying and characterizing alternative splicing (AS) enables our understanding of the biological role of transcript isoform diversity. This study describes the use of publicly available RNA-Seq data to identify and characterize the global diversity of AS isoforms in maize using the inbred lines B73 and Mo17, and a related species, sorghum. Identification and characterization of AS within maize tissues revealed that genes expressed in seed exhibit the largest differential AS relative to other tissues examined. Additionally, differences in AS between the two genotypes B73 and Mo17 are greatest within genes expressed in seed. We demonstrate that changes in the level of alternatively spliced transcripts (intron retention and exon skipping) do not solely reflect differences in total transcript abundance, and we present evidence that intron retention may act to fine-tune gene expression across seed development stages. Furthermore, we have identified temperature sensitive AS in maize and demonstrate that drought-induced changes in AS involve distinct sets of genes in reproductive and vegetative tissues. Examining our identified AS isoforms within B73 × Mo17 recombinant inbred lines (RILs) identified splicing QTL (sQTL). The 43.3% of cis-sQTL regulated junctions are actually identified as alternatively spliced junctions in our analysis, while 10 Mb windows on each side of 48.2% of trans-sQTLs overlap with splicing related genes. Using sorghum as an out-group enabled direct examination of loss or conservation of AS between homeologous genes representing the two subgenomes of maize. We identify several instances where AS isoforms that are conserved between one maize homeolog and its sorghum ortholog are absent from the second maize homeolog, suggesting that these AS isoforms may have been lost after the maize whole genome duplication event. This comprehensive analysis provides new insights into the complexity of AS in maize.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.