Maintenance and expansion of genetic and trait variation following domestication in a clonal crop.

Clonal propagation enables favourable crop genotypes to be rapidly selected and multiplied. However, the absence of sexual propagation can lead to low genetic diversity and accumulation of deleterious mutations, which may eventually render crops less resilient to pathogens or environmental change. To better understand this trade-off, we characterize the domestication and contemporary genetic diversity of Enset (Ensete ventricosum), an indigenous African relative of bananas (Musa) and a principal starch staple for 20 million Ethiopians. Wild enset reproduction occurs strictly by sexual outcrossing, but for cultivation, it is propagated clonally and associated with diversification and specialization into hundreds of named landraces. We applied tGBS sequencing to generate genome-wide genotypes for 192 accessions from across enset's cultivated distribution, and surveyed 1340 farmers on enset agronomic traits. Overall, reduced heterozygosity in the domesticated lineage was consistent with a domestication bottleneck that retained 37% of wild diversity. However, an excess of putatively deleterious missense mutations at low frequency present as heterozygotes suggested an accumulation of mutational load in clonal domesticated lineages. Our evidence indicates that the major domesticated lineages initially arose through historic sexual recombination associated with a domestication bottleneck, followed by the amplification of favourable genotypes through an extended period of clonal propagation. Among domesticated lineages, we found a significant phylogenetic signal for multiple farmer-identified food, nutrition and disease resistance traits and little evidence of contemporary recombination. The development of future-climate adapted genotypes may require crop breeding, but outcrossing risks exposing deleterious alleles as homozygotes. This trade-off may partly explain the ubiquity and persistence of clonal propagation over recent centuries of comparative climate stability.

Assembly-free quantification of vagrant DNA inserts.

Inserts of DNA from extranuclear sources, such as organelles and microbes, are common in eukaryote nuclear genomes. However, sequence similarity between the nuclear and extranuclear DNA, and a history of multiple insertions, make the assembly of these regions challenging. Consequently, the number, sequence and location of these vagrant DNAs cannot be reliably inferred from the genome assemblies of most organisms. We introduce two statistical methods to estimate the abundance of nuclear inserts even in the absence of a nuclear genome assembly. The first (intercept method) only requires low-coverage (<1×) sequencing data, as commonly generated for population studies of organellar and ribosomal DNAs. The second method additionally requires that a subset of the individuals carry extranuclear DNA with diverged genotypes. We validated our intercept method using simulations and by re-estimating the frequency of human NUMTs (nuclear mitochondrial inserts). We then applied it to the grasshopper Podisma pedestris, exceptional for both its large genome size and reports of numerous NUMT inserts, estimating that NUMTs make up 0.056% of the nuclear genome, equivalent to >500 times the mitochondrial genome size. We also re-analysed a museomics data set of the parrot Psephotellus varius, obtaining an estimate of only 0.0043%, in line with reports from other species of bird. Our study demonstrates the utility of low-coverage high-throughput sequencing data for the quantification of nuclear vagrant DNAs. Beyond quantifying organellar inserts, these methods could also be used on endosymbiont-derived sequences. We provide an R implementation of our methods called "vagrantDNA" and code to simulate test data sets.

The effect of curricular reform on gross anatomy laboratory examination performance: An institutional analysis.

Many medical schools have undergone curricular reform recently. With these reforms, time spent teaching anatomy has been reduced, and there has been a general shift to a pass/fail grading system. At Indiana University School of Medicine (IUSM), a new curriculum was implemented in fall 2016. The year-long human gross anatomy course taught in 2015 was condensed into an integrated, semester-long course starting in 2016. Additionally, the grading scale shifted to pass/fail. This study examined first-year medical student performance on anatomy practical laboratory examinations-specifically, among lower-order (pure identification) questions and higher-order (function, innervation) questions. Participants included medical students from a pre-curricular reform cohort (year 2015, 34 students) and two post-curricular reform cohorts (years 2016, 30 students and 2017, 33 students). A Kruskal-Wallis ANOVA test was used to determine differences of these questions among the three cohorts. Additionally, 40 of the same lower-order questions that were asked on gross anatomy laboratory examinations from medical student cohort year 2015 and year 2016 were further analyzed using an independent samples t-test. Results demonstrated that the pre-curricular reform cohort scored significantly higher on both lower-order (median = 81, p < 0.001) and higher-order questions (median = 82.5, p < 0.05) than both post-curricular reform cohorts. Additionally, when reviewing the selected 40 similar questions, it was found that the pre-curricular reform cohort averaged significantly higher (82.1 ± 16.1) than the post-curricular reform cohort from 2016 (69.3 ± 21.8, p = 0.004). This study provides evidence about the impact of curricular reform on medical student anatomical knowledge.

Non-Lethal Detection of Frog Virus 3-Like (RUK13) and Common Midwife Toad Virus-Like (PDE18) Ranaviruses in Two UK-Native Amphibian Species.

Ranaviruses have been involved in amphibian mass mortality events worldwide. Effective screening to control this pathogen is essential; however, current sampling methods are unsuitable for the detection of subclinical infections. Non-lethal screening is needed to prevent both further spread of ranavirus and losses of at-risk species. To assess non-lethal sampling methods, we conducted two experiments: bath exposing common frogs to RUK13 ranavirus at three concentrations, and exposing common toads to RUK13 or PDE18. Non-lethal sampling included buccal, digit, body and tank swabs, along with toe clips and stool taken across three time-points post-exposure. The presence/load of ranavirus was examined using quantitative PCR in 11 different tissues obtained from the same euthanised animals (incl. liver, gastro-intestinal tract and kidney). Buccal swab screening had the highest virus detection rate in both species (62% frogs; 71% toads) and produced consistently high virus levels compared to other non-lethal assays. The buccal swab was effective across multiple stages of infection and differing infection intensities, though low levels of infection were more difficult to detect. Buccal swab assays competed with, and even outperformed, lethal sampling in frogs and toads, respectively. Successful virus detection in the absence of clinical signs was observed (33% frogs; 50% toads); we found no difference in detectability for RUK13 and PDE18. Our results suggest that buccal swabbing could replace lethal sampling for screening and be introduced as standard practice for ranavirus surveillance.

Spatially clustered count data provide more efficient search strategies in invasion biology and disease control.

Geographic profiling, a mathematical model originally developed in criminology, is increasingly being used in ecology and epidemiology. Geographic profiling boasts a wide range of applications, such as finding source populations of invasive species or breeding sites of vectors of infectious disease. The model provides a cost-effective approach for prioritizing search strategies for source locations and does so via simple data in the form of the positions of each observation, such as individual sightings of invasive species or cases of a disease. In doing so, however, classic geographic profiling approaches fail to make the distinction between those areas containing observed absences and those areas where no data were recorded. Absence data are generated via spatial sampling protocols but are often discarded during the inference process. Here we construct a geographic profiling model that resolves these issues by making inferences via count data, analyzing a set of discrete sentinel locations at which the number of encounters has been recorded. Crucially, in our model this number can be zero. We verify the ability of this new model to estimate source locations and other parameters of practical interest via a Bayesian power analysis. We also measure model performance via real-world data in which the model infers breeding locations of mosquitoes in bromeliads in Miami-Dade County, Florida, USA. In both cases, our novel model produces more efficient search strategies by shifting focus from those areas containing observed absences to those with no data, an improvement over existing models that treat these areas equally. Our model makes important improvements upon classic geographic profiling methods, which will significantly enhance real-world efforts to develop conservation management plans and targeted interventions.

Genomic architecture and evolutionary antagonism drive allelic expression bias in the social supergene of red fire ants.

Supergene regions maintain alleles of multiple genes in tight linkage through suppressed recombination. Despite their importance in determining complex phenotypes, our empirical understanding of early supergene evolution is limited. Here we focus on the young 'social' supergene of fire ants, a powerful system for disentangling the effects of evolutionary antagonism and suppressed recombination. We hypothesize that gene degeneration and social antagonism shaped the evolution of the fire ant supergene, resulting in distinct patterns of gene expression. We test these ideas by identifying allelic differences between supergene variants, characterizing allelic expression across populations, castes and body parts, and contrasting allelic expression biases with differences in expression between social forms. We find strong signatures of gene degeneration and gene-specific dosage compensation. On this background, a small portion of the genes has the signature of adaptive responses to evolutionary antagonism between social forms.

Red fire ants (Solenopsis invicta) are native to South America, but the species has spread to North America, Australia and New Zealand where it can be an invasive pest. A reason for this species' invasiveness types of colonies : one with a single egg-laying queen and another with several queens. However, it is not possible to simply add more queens to a colony with one queen. Instead, the number of queens in a colony is controlled genetically, by a chromosome known as the 'social chromosome'. Like many other animals, red fire ants are diploid: their cells have two copies of each chromosome, which can carry two different versions of each gene. The social chromosome is no different, and it comes in two variants, SB and Sb. Each ant can therefore have either two SB chromosomes, leading to a colony with a single queen; or one SB chromosome and one Sb chromosome, leading to a colony with multiple queens. Ants with two copies of the Sb variant die when they are young, so the Sb version is inherited in a similar way to how the Y chromosome is passed on in humans. However, the social chromosome in red fire ants appeared less than one million years ago, making it much younger than the human Y chromosome, which is 180 million years old. This makes the social chromosome a good candidate for examining the early evolution of special chromosome variants that are only inherited. How differences between the SB and the Sb chromosomes are evolving is an open question, however. Perhaps each version of the social chromosome has been optimised through natural selection to one colony type. Another suggestion is that the Sb chromosome has degenerated over time because its genes cannot be 'reshuffled' as they would be on normal chromosomes. Martinez-Ruiz et al. compared genetic variants on the SB and Sb chromosomes, along with their expression in different types of ant colonies. The analysis showed that the Sb variant is in fact breaking down because of the lack of gene shuffling. This loss is compensated by intact copies of the same genes found on the SB variant, which explains why ants with the Sb variant can only survive if they also carry the SB version. Only a handful of genes on the social chromosomes appear to have been optimised by natural selection. Therefore Martinez-Ruiz et al. concluded the differences between the two chromosomes that lead to different colony types are collateral effects of Sb's inability to reshuffle its genes. This work reveals how a special chromosome similar to the Y chromosome in humans evolved. It also shows how multiple complex evolutionary forces can shape a species' genetic makeup and social forms.

Genomic assessment of local adaptation in dwarf birch to inform assisted gene flow.

When populations of a rare species are small, isolated and declining under climate change, some populations may become locally maladapted. Detecting this maladaptation may allow effective rapid conservation interventions, even if based on incomplete knowledge. Population maladaptation may be estimated by finding genome-environment associations (GEA) between allele frequencies and environmental variables across a local species range, and identifying populations whose allele frequencies do not fit with these trends. We can then design assisted gene flow strategies for maladapted populations, to adjust their allele frequencies, entailing lower levels of intervention than with undirected conservation action. Here, we investigate this strategy in Scottish populations of the montane plant dwarf birch (Betula nana). In genome-wide restriction site-associated single nucleotide polymorphism (SNP) data, we found 267 significant associations between SNP loci and environmental variables. We ranked populations by maladaptation estimated using allele frequency deviation from the general trends at these loci; this gave a different prioritization for conservation action than the Shapely Index, which seeks to preserve rare neutral variation. Populations estimated to be maladapted in their allele frequencies at loci associated with annual mean temperature were found to have reduced catkin production. Using an environmental niche modelling (ENM) approach, we found annual mean temperature (35%), and mean diurnal range (15%), to be important predictors of the dwarf birch distribution. Intriguingly, there was a significant correlation between the number of loci associated with each environmental variable in the GEA and the importance of that variable in the ENM. Together, these results suggest that the same environmental variables determine both adaptive genetic variation and species range in Scottish dwarf birch. We suggest an assisted gene flow strategy that aims to maximize the local adaptation of dwarf birch populations under climate change by matching allele frequencies to current and future environments.

Genomic basis of European ash tree resistance to ash dieback fungus.

Populations of European ash trees (Fraxinus excelsior) are being devastated by the invasive alien fungus Hymenoscyphus fraxineus, which causes ash dieback. We sequenced whole genomic DNA from 1,250 ash trees in 31 DNA pools, each pool containing trees with the same ash dieback damage status in a screening trial and from the same seed-source zone. A genome-wide association study identified 3,149 single nucleotide polymorphisms (SNPs) associated with low versus high ash dieback damage. Sixty-one of the 192 most significant SNPs were in, or close to, genes with putative homologues already known to be involved in pathogen responses in other plant species. We also used the pooled sequence data to train a genomic prediction model, cross-validated using individual whole genome sequence data generated for 75 healthy and 75 damaged trees from a single seed source. The model's genomic estimated breeding values (GEBVs) allocated these 150 trees to their observed health statuses with 67% accuracy using 10,000 SNPs. Using the top 20% of GEBVs from just 200 SNPs, we could predict observed tree health with over 90% accuracy. We infer that ash dieback resistance in F. excelsior is a polygenic trait that should respond well to both natural selection and breeding, which could be accelerated using genomic prediction.

Modelling Ranavirus Transmission in Populations of Common Frogs (Rana temporaria) in the United Kingdom.

Ranaviruses began emerging in common frogs (Rana temporaria) in the United Kingdom in the late 1980s and early 1990s, causing severe disease and declines in the populations of these animals. Herein, we explored the transmission dynamics of the ranavirus(es) present in common frog populations, in the context of a simple susceptible-infected (SI) model, using parameters derived from the literature. We explored the effects of disease-induced population decline on the dynamics of the ranavirus. We then extended the model to consider the infection dynamics in populations exposed to both ulcerative and hemorrhagic forms of the ranaviral disease. The preliminary investigation indicated the important interactions between the forms. When the ulcerative form was present in a population and the hemorrhagic form was later introduced, the hemorrhagic form of the disease needed to be highly contagious, to persist. We highlighted the areas where further research and experimental evidence is needed and hope that these models would act as a guide for further research into the amphibian disease dynamics.

Effects of historic and projected climate change on the range and impacts of an emerging wildlife disease.

The global trend of increasing environmental temperatures is often predicted to result in more severe disease epidemics. However, unambiguous evidence that temperature is a driver of epidemics is largely lacking, because it is demanding to demonstrate its role among the complex interactions between hosts, pathogens, and their shared environment. Here, we apply a three-pronged approach to understand the effects of temperature on ranavirus epidemics in UK common frogs, combining in vitro, in vivo, and field studies. Each approach suggests that higher temperatures drive increasing severity of epidemics. In wild populations, ranavirosis incidents were more frequent and more severe at higher temperatures, and their frequency increased through a period of historic warming in the 1990s. Laboratory experiments using cell culture and whole animal models showed that higher temperature increased ranavirus propagation, disease incidence, and mortality rate. These results, combined with climate projections, predict severe ranavirosis outbreaks will occur over wider areas and an extended season, possibly affecting larval recruitment. Since ranaviruses affect a variety of ectothermic hosts (amphibians, reptiles, and fish), wider ecological damage could occur. Our three complementary lines of evidence present a clear case for direct environmental modulation of these epidemics and suggest management options to protect species from disease.

Genome-wide haplotype analysis improves trait predictions in Brassica napus hybrids.

Combining ability is crucial for parent selection in crop hybrid breeding. Many studies have attempted to provide reliable and quick methods to identify genome regions in parental lines correlating with improved hybrid performance. The local haplotype patterns surrounding densely spaced DNA markers include a large amount of genetic information, and analysis of the relationships between haplotypes and hybrid performance can provide insight into the underlying genome regions which might contribute to enhancing combining ability. Here, we generated 24,403 single-copy, genome-wide SNP loci and calculated the general combining ability (GCA) of 950 hybrids from a diverse panel of 475 pollinators of spring-type canola inbred lines crossed with two testers for days to flowering (DTF) and seed glucosinolate content (GSL). We performed a genome-wide analysis of the haplotypes and detected eight and seven haplotype regions that were significantly associated with the GCA values for DTF and seed GSL, respectively. Additionally, two haplotype blocks containing orthologs of flowering time genes FLOWERING LOCUS T (FT) and FLOWERING LOCUS C (FLC) on chromosome A02 showed additive epistatic interactions influencing flowering time. Moreover, two homoeologous haplotype regions on chromosomes A02 and C02 corresponded to major quantitative trait loci (QTL) for GSL which showed additive effects related to reduction of seed GSL in F1 hybrids. Our study showed that haplotype analysis has the potential to substantially improve the efficiency of hybrid breeding programs.

Interactions between plant genome size, nutrients and herbivory by rabbits, molluscs and insects on a temperate grassland.

Angiosperm genome sizes (GS) vary ca 2400-fold. Recent research has shown that GS influences plant abundance, and plant competition. There are also tantalizing reports that herbivores may select plants as food dependent on their GS. To test the hypothesis that GS plays a role in shaping plant communities under herbivore pressure, we exploit a grassland experiment that has experimentally excluded herbivores and applied nutrient over 8 years. Using phylogenetically informed statistical models and path analyses, we show that under rabbit grazing, plant species with small GS generated the most biomass. By contrast, on mollusc and insect-grazed plots, it was the plant species with larger GS that increased in biomass. GS was also shown to influence plant community properties (e.g. competitive strategy, total biomass) although the impact varied between different herbivore guilds (i.e. rabbits versus invertebrates) and nutrient inputs. Overall, we demonstrate that GS plays a role in influencing plant-herbivore interactions, and suggest potential reasons for this response, which include the impact of GS on a plant's response to different herbivore guilds, and on a plant's nutrient quality. The inclusion of GS in ecological models has the potential to expand our understanding of plant productivity and community ecology under nutrient and herbivore stress.

Genetic diversity maintained among fragmented populations of a tree undergoing range contraction.

Dwarf birch (Betula nana) has a widespread boreal distribution but has declined significantly in Britain where populations are now highly fragmented. We analyzed the genetic diversity of these fragmented populations using markers that differ in mutation rate: conventional microsatellites markers (PCR-SSRs), RADseq generated transition and transversion SNPs (RAD-SNPs), and microsatellite markers mined from RADseq reads (RAD-SSRs). We estimated the current population sizes by census and indirectly, from the linkage-disequilibrium found in the genetic surveys. The two types of estimate were highly correlated. Overall, we found genetic diversity to be only slightly lower in Britain than across a comparable area in Scandinavia where populations are large and continuous. While the ensemble of British fragments maintain diversity levels close to Scandinavian populations, individually they have drifted apart and lost diversity; particularly the smaller populations. An ABC analysis, based on coalescent models, favors demographic scenarios in which Britain maintained high levels of genetic diversity through post-glacial re-colonization. This diversity has subsequently been partitioned into population fragments that have recently lost diversity at a rate corresponding to the current population-size estimates. We conclude that the British population fragments retain sufficient genetic resources to be the basis of conservation and re-planting programmes. Use of markers with different mutation rates gives us greater confidence and insight than one marker set could have alone, and we suggest that RAD-SSRs are particularly useful as high mutation-rate marker set with a well-specified ascertainment bias, which are widely available yet often neglected in existing RAD datasets.

The fire ant social chromosome supergene variant Sb shows low diversity but high divergence from SB.

Variation in social behaviour is common, yet little is known about the genetic architectures underpinning its evolution. A rare exception is in the fire ant Solenopsis invicta: Alternative variants of a supergene region determine whether a colony will have exactly one or up to dozens of queens. The two variants of this region are carried by a pair of 'social chromosomes', SB and Sb, which resemble a pair of sex chromosomes. Recombination is suppressed between the two chromosomes in the supergene region. While the X-like SB can recombine with itself in SB/SB queens, recombination is effectively absent in the Y-like Sb because Sb/Sb queens die before reproducing. Here, we analyse whole-genome sequences of eight haploid SB males and eight haploid Sb males. We find extensive SB-Sb differentiation throughout the >19-Mb-long supergene region. We find no evidence of 'evolutionary strata' with different levels of divergence comparable to those reported in several sex chromosomes. A high proportion of substitutions between the SB and Sb haplotypes are nonsynonymous, suggesting inefficacy of purifying selection in Sb sequences, similar to that for Y-linked sequences in XY systems. Finally, we show that the Sb haplotype of the supergene region has 635-fold less nucleotide diversity than the rest of the genome. We discuss how this reduction could be due to a recent selective sweep affecting Sb specifically or associated with a population bottleneck during the invasion of North America by the sampled population.

Reconstructing the emergence of a lethal infectious disease of wildlife supports a key role for spread through translocations by humans.

There have been few reconstructions of wildlife disease emergences, despite their extensive impact on biodiversity and human health. This is in large part attributable to the lack of structured and robust spatio-temporal datasets. We overcame logistical problems of obtaining suitable information by using data from a citizen science project and formulating spatio-temporal models of the spread of a wildlife pathogen (genus Ranavirus, infecting amphibians). We evaluated three main hypotheses for the rapid increase in disease reports in the UK: that outbreaks were being reported more frequently, that climate change had altered the interaction between hosts and a previously widespread pathogen, and that disease was emerging due to spatial spread of a novel pathogen. Our analysis characterized localized spread from nearby ponds, consistent with amphibian dispersal, but also revealed a highly significant trend for elevated rates of additional outbreaks in localities with higher human population density-pointing to human activities in also spreading the virus. Phylogenetic analyses of pathogen genomes support the inference of at least two independent introductions into the UK. Together these results point strongly to humans repeatedly translocating ranaviruses into the UK from other countries and between UK ponds, and therefore suggest potential control measures.

Estimating the Number of Subpopulations (K) in Structured Populations.

A key quantity in the analysis of structured populations is the parameter K, which describes the number of subpopulations that make up the total population. Inference of K ideally proceeds via the model evidence, which is equivalent to the likelihood of the model. However, the evidence in favor of a particular value of K cannot usually be computed exactly, and instead programs such as Structure make use of heuristic estimators to approximate this quantity. We show-using simulated data sets small enough that the true evidence can be computed exactly-that these heuristics often fail to estimate the true evidence and that this can lead to incorrect conclusions about K Our proposed solution is to use thermodynamic integration (TI) to estimate the model evidence. After outlining the TI methodology we demonstrate the effectiveness of this approach, using a range of simulated data sets. We find that TI can be used to obtain estimates of the model evidence that are more accurate and precise than those based on heuristics. Furthermore, estimates of K based on these values are found to be more reliable than those based on a suite of model comparison statistics. Finally, we test our solution in a reanalysis of a white-footed mouse data set. The TI methodology is implemented for models both with and without admixture in the software MavericK1.0.

Unidirectional diploid-tetraploid introgression among British birch trees with shifting ranges shown by restriction site-associated markers.

Hybridization may lead to introgression of genes among species. Introgression may be bidirectional or unidirectional, depending on factors such as the demography of the hybridizing species, or the nature of reproductive barriers between them. Previous microsatellite studies suggested bidirectional introgression between diploid Betula nana (dwarf birch) and tetraploid B. pubescens (downy birch) and also between B. pubescens and diploid B. pendula (silver birch) in Britain. Here, we analyse introgression among these species using 51 237 variants in restriction site-associated (RAD) markers in 194 individuals, called with allele dosages in the tetraploids. In contrast to the microsatellite study, we found unidirectional introgression into B. pubescens from both of the diploid species. This pattern fits better with the expected nature of the reproductive barrier between diploids and tetraploids. As in the microsatellite study, introgression into B. pubescens showed clear clines with increasing introgression from B. nana in the north and from B. pendula in the south. Unlike B. pendula alleles, introgression of B. nana alleles was found far from the current area of sympatry or allopatry between B. nana and B. pubescens. This pattern fits a shifting zone of hybridization due to Holocene reduction in the range of B. nana and expansion in the range of B. pubescens.

Genome size and ploidy influence angiosperm species' biomass under nitrogen and phosphorus limitation.

Angiosperm genome sizes (GS) range c. 2400-fold, and as nucleic acids are amongst the most phosphorus- (P) and nitrogen (N)-demanding cellular biomolecules, we test the hypothesis that a key influence on plant biomass and species composition is the interaction between N and P availability and plant GS. We analysed the impact of different nutrient regimes on above-ground biomass of angiosperm species with different GS, ploidy level and Grime's C-S-R (competitive, stress-tolerant, ruderal) plant strategies growing at the Park Grass Experiment (Rothamsted, UK), established in 1856. The biomass-weighted mean GS of species growing on plots with the addition of both N and P fertilizer were significantly higher than that of plants growing on control plots and plots with either N or P. The plants on these N + P plots are dominated by polyploids with large GS and a competitive plant strategy. The results are consistent with our hypothesis that large genomes are costly to build and maintain under N and P limitation. Hence GS and ploidy are significant traits affecting biomass growth under different nutrient regimes, influencing plant community composition and ecosystem dynamics. We propose that GS is a critical factor needed in models that bridge the knowledge gap between biodiversity and ecosystem functioning.

Genomic Prediction of Testcross Performance in Canola (Brassica napus).

Genomic selection (GS) is a modern breeding approach where genome-wide single-nucleotide polymorphism (SNP) marker profiles are simultaneously used to estimate performance of untested genotypes. In this study, the potential of genomic selection methods to predict testcross performance for hybrid canola breeding was applied for various agronomic traits based on genome-wide marker profiles. A total of 475 genetically diverse spring-type canola pollinator lines were genotyped at 24,403 single-copy, genome-wide SNP loci. In parallel, the 950 F1 testcross combinations between the pollinators and two representative testers were evaluated for a number of important agronomic traits including seedling emergence, days to flowering, lodging, oil yield and seed yield along with essential seed quality characters including seed oil content and seed glucosinolate content. A ridge-regression best linear unbiased prediction (RR-BLUP) model was applied in combination with 500 cross-validations for each trait to predict testcross performance, both across the whole population as well as within individual subpopulations or clusters, based solely on SNP profiles. Subpopulations were determined using multidimensional scaling and K-means clustering. Genomic prediction accuracy across the whole population was highest for seed oil content (0.81) followed by oil yield (0.75) and lowest for seedling emergence (0.29). For seed yieId, seed glucosinolate, lodging resistance and days to onset of flowering (DTF), prediction accuracies were 0.45, 0.61, 0.39 and 0.56, respectively. Prediction accuracies could be increased for some traits by treating subpopulations separately; a strategy which only led to moderate improvements for some traits with low heritability, like seedling emergence. No useful or consistent increase in accuracy was obtained by inclusion of a population substructure covariate in the model. Testcross performance prediction using genome-wide SNP markers shows considerable potential for pre-selection of promising hybrid combinations prior to resource-intensive field testing over multiple locations and years.