Hybrid Fitness Research Articles

Ecological divergence in sympatry causes gene misexpression in hybrids.

Ecological speciation occurs when reproductive isolation evolves as a byproduct of adaptive divergence between populations. Selection favouring gene regulatory divergence between species could result in transgressive levels of gene expression in F1 hybrids that may lower hybrid fitness. We combined 58 resequenced genomes with 124 transcriptomes to identify patterns of hybrid gene misexpression that may be driven by adaptive regulatory divergence within a young radiation of Cyprinodon pupfishes, which consists of a dietary generalist and two trophic specialists-a molluscivore and a scale-eater. We found more differential gene expression between closely related sympatric specialists than between allopatric generalist populations separated by 1,000km. Intriguingly, 9.6% of genes that were differentially expressed between sympatric species were also misexpressed in F1 hybrids. A subset of these genes were in highly differentiated genomic regions and enriched for functions important for trophic specialization, including head, muscle and brain development. These regions also included genes that showed evidence of hard selective sweeps and were significantly associated with oral jaw length-the most rapidly diversifying skeletal trait in this radiation. Our results indicate that divergent ecological selection in sympatry can contribute to hybrid gene misexpression which may act as a reproductive barrier between nascent species.

Detecting differences in the topology of scale-free networks grown under time-dynamic topological fitness

The fitness model was introduced in the literature to expand the Barabasi-Albert model’s generative mechanism, which produces scale-free networks under the control of degree. However, the fitness model has not yet been studied in a comprehensive context because most models are built on invariant fitness as the network grows and time-dynamics mainly concern new nodes joining the network. This mainly static consideration restricts fitness in generating scale-free networks only when the underlying fitness distribution is power-law, a fact which makes the hybrid fitness models based on degree-driven preferential attachment to remain the most attractive models in the literature. This paper advances the time-dynamic conceptualization of fitness, by studying scale-free networks generated under topological fitness that changes as the network grows, where the fitness is controlled by degree, clustering coefficient, betweenness, closeness, and eigenvector centrality. The analysis shows that growth under time-dynamic topological fitness is indifferent to the underlying fitness distribution and that different topological fitness generates networks of different topological attributes, ranging from a mesh-like to a superstar-like pattern. The results also show that networks grown under the control of betweenness centrality outperform the other networks in scale-freeness and the majority of the other topological attributes. Overall, this paper contributes to broadening the conceptualization of fitness to a more time-dynamic context.

The contribution of multiple barriers to reproduction between edaphically divergent lineages in the Amazonian tree Protium subserratum (Burseraceae).

Disentangling the strength and importance of barriers to reproduction that arise between diverging lineages is central to our understanding of species origin and maintenance. To date, the vast majority of studies investigating the importance of different barriers to reproduction in plants have focused on short‐lived temperate taxa while studies of reproductive isolation in trees and tropical taxa are rare. Here, we systematically examine multiple barriers to reproduction in an Amazonian tree, Protium subserratum (Burseraceae) with diverging lineages of soil specialist ecotypes. Using observational, molecular, distributional, and experimental data, we aimed to quantify the contributions of individual prezygotic and postzygotic barriers including ecogeographic isolation, flowering phenology, pollinator assemblage, pollen adhesion, pollen germination, pollen tube growth, seed development, and hybrid fitness to total reproductive isolation between the ecotypes. We were able to identify five potential barriers to reproduction including ecogeographic isolation, phenological differences, differences in pollinator assemblages, differential pollen adhesion, and low levels of hybrid seed development. We demonstrate that ecogeographic isolation is a strong and that a combination of intrinsic and extrinsic prezygotic and postzygotic barriers may be acting to maintain near complete reproductive isolation between edaphically divergent populations of the tropical tree, P. subserratum.

Assortative Mating in Hybrid Zones Is Remarkably Ineffective in Promoting Speciation.

Partial prezygotic isolation is often viewed as more important than partial postzygotic isolation (low fitness of hybrids) early in the process of speciation. I simulate secondary contact between two populations (species) to examine effects of assortative mating and low hybrid fitness in preventing blending. A small reduction in hybrid fitness (e.g., by 10%) produces a narrower hybrid zone than a strong but imperfect mating preference (e.g., 10 times stronger preference for conspecific over heterospecific mates). In the latter case, rare F1 hybrids find each other attractive (due to assortative mating), leading to the buildup of a continuum of intermediates. The weakness of assortative mating compared with reduced fitness of hybrids in preventing blending is robust to varying genetic bases of these traits. Assortative mating is most powerful in limiting blending when it is encoded by a single locus or is essentially complete, or when there is a large mate search cost. In these cases assortative mating is likely to cause hybrids to have low fitness, due to frequency-dependent mating disadvantage of individuals of rare mating types. These results prompt a questioning of the concept of partial prezygotic isolation, since it is not very isolating unless there is also postzygotic isolation.

Computer Simulation Model of a Hybrid Zone: Assortative Mating in the Role of Selection against Hybrids

The assumption of low hybrid fitness serves as the cornerstone in the logical justification of the discreteness of species in terms of the biological concept. The majority of natural hybrid zones are narrow and stable in time; thus, this proves, as mathematical models have shown, the existence of selection against hybrids. It is assumed that, since the coadaptive gene complexes of the parent forms are destroyed in hybrids, they must be less viable and/or fertile. However, this statement remains no more than a hypothesis. The low reproductive success of hybrids may have other causes, e.g., they may have difficulties in pair formation because of the rarity and strangeness of their appearance. A computer simulation model has been constructed to verify this assumption. The main purpose of this work is to give answers to the following questions. (1) Is it possible to form a narrow and stable hybrid zone with assortative mating but without lower viability and fertility of the hybrids? Assortative mating in the model is due to sexual imprinting: descendants remember the image of one of their parents and seek an appropriate partner. (2) Is it possible to complete the speciation process under these conditions and what is necessary for this? Here, we interpret speciation as the process of the establishment of sympatry between the initially allopatric forms with the maintenance of their appearance and a low level of hybridization. The model study gave positive answers for both questions. Without biotopic segregation and with assortative mating, the hybrid zone is narrow and stable in time. Hybrids remain unmated more often than parent forms (5–8 and 1–3% respectively), which results in selection pressure in favor of a narrow hybrid zone. In addition to assortative mating, biotopic segregation is necessary to establish sympatry, and hybrids have to be less successful than parent forms in the occupation of an individual territory. In opposite case, when the biotopes of the both parent forms are optimal for hybrids, the contact zone became wider at first and then, after the smoothing of biotopic segregation, they started to slowly shrink to a narrow hybrid zone. As a result, the two forms became indistinguishable by selectively neutral loci (DNA markers analog) over the entire space of the model.

Hybridization and transgressive exploration of colour pattern and wing morphology in Heliconius butterflies.

Hybridization can generate novel phenotypes distinct from those of parental lineages, a phenomenon known as transgressive trait variation. Transgressive phenotypes might negatively or positively affect hybrid fitness, and increase available variation. Closely related species of Heliconius butterflies regularly produce hybrids in nature, and hybridization is thought to play a role in the diversification of novel wing colour patterns despite strong stabilizing selection due to interspecific mimicry. Here, we studied wing phenotypes in first- and second-generation hybrids produced by controlled crosses between either two co-mimetic species of Heliconius or between two nonmimetic species. We quantified wing size, shape and colour pattern variation and asked whether hybrids displayed transgressive wing phenotypes. Discrete traits underlain by major-effect loci, such as the presence or absence of colour patches, generate novel phenotypes. For quantitative traits, such as wing shape or subtle colour pattern characters, hybrids only exceed the parental range in specific dimensions of the morphological space. Overall, our study addresses some of the challenges in defining and measuring phenotypic transgression for multivariate traits and our data suggest that the extent to which transgressive trait variation in hybrids contributes to phenotypic diversity depends on the complexity and the genetic architecture of the traits.

Cytonuclear Genetic Incompatibilities in Plant Speciation.

Due to the endosymbiotic origin of organelles, a pattern of coevolution and coadaptation between organellar and nuclear genomes is required for proper cell function. In this review, we focus on the impact of cytonuclear interaction on the reproductive isolation of plant species. We give examples of cases where species exhibit barriers to reproduction which involve plastid-nuclear or mito-nuclear genetic incompatibilities, and describe the evolutionary processes at play. We also discuss potential mechanisms of hybrid fitness recovery such as paternal leakage. Finally, we point out the possible interplay between plant mating systems and cytonuclear coevolution, and its consequence on plant speciation.

Plasticity of Mitochondrial DNA Inheritance and its Impact on Nuclear Gene Transcription in Yeast Hybrids.

Mitochondrial DNA (mtDNA) in yeast is biparentally inherited, but colonies rapidly lose one type of parental mtDNA, thus becoming homoplasmic. Therefore, hybrids between the yeast species possess two homologous nuclear genomes, but only one type of mitochondrial DNA. We hypothesise that the choice of mtDNA retention is influenced by its contribution to hybrid fitness in different environments, and the allelic expression of the two nuclear sub-genomes is affected by the presence of different mtDNAs in hybrids. Saccharomyces cerevisiae/S. uvarum hybrids preferentially retained S. uvarum mtDNA when formed on rich media at colder temperatures, while S. cerevisiae mtDNA was primarily retained on non-fermentable carbon source, at any temperature. Transcriptome data for hybrids harbouring different mtDNA showed a strong environmentally dependent allele preference, which was more important in respiratory conditions. Co-expression analysis for specific biological functions revealed a clear pattern of concerted allelic transcription within the same allele type, which supports the notion that the hybrid cell works preferentially with one set of parental alleles (or the other) for different cellular functions. Given that the type of mtDNA retained in hybrids affects both nuclear expression and fitness, it might play a role in driving hybrid genome evolution in terms of gene retention and loss.

Modelling population dynamics based on experimental trials with genetically modified (RIDL) mosquitoes

Recently, the RIDL-SIT technology has been field-tested for control of Aedes aegypti. The technique consists of releasing genetically modified mosquitoes carrying a “lethal gene”. In 2016 the World Health Organization (WHO) and the Pan-American Health Organization (PAHO) recommended to their constituent countries to test the new technologies proposed to control Aedes aegypti populations. However, issues concerning effectiveness and ecological impact have not been thoroughly studied so far. In order to study these issues, we develop an ecological model. It presents interdependent dynamics of mosquito populations and food in a homogeneous setting. Mosquito populations are described in a stochastic compartmental setup, in terms of reaction norms depending on the available food in the environment. The development of the model allows us to indicate some critical biological knowledge that is missing and could (should) be produced. Hybridisation levels, release numbers during and after intervention and population recovery time after the intervention as a function of intervention duration and target are calculated under different hypotheses with regard to the fitness of hybrids and compared with two field studies of actual interventions. This minimal model should serve as a basis for detailed models when the necessary information to construct them is produced. For the time being, the model shows that nature will not clean non-lethal introgressed genes.

Speciation Associated with Shifts in Migratory Behavior in an Avian Radiation.

Migratory animals move up to thousands of kilometers every year [1]. Losses of migration (i.e., migratory drop-offs) occur when individuals of a migratory species stop migrating and establish founder sedentary populations, a phenomenon documented in birds [2-5] and butterflies [6]. In theory, losses-and also gains-of migration might promote speciation if sedentary and migratory populations become reproductively isolated [7-9]. Because migratory and sedentary strategies involve alternative physiological, behavioral, and morphological traits [10-13], divergence along multiple axes of organismal function is expected to accompany switches in migratory behavior, potentially accelerating speciation. We present evidence of speciation driven by a migratory drop-off in the fork-tailed flycatcher (Tyrannus savana) resulting in reproductive isolation likely driven by changes in breeding schedules (allochronic speciation [13-15]) and geographic isolation of breeding grounds (allopatric speciation [16]). Phylogenetic analyses across New World flycatchers (Tyrannidae) showed that an association between speciation and drop-offs is also observable at a macroevolutionary scale. Loss of migration was significantly more frequent than its gain, and speciation rates of migratory and partially migratory lineages (i.e., species having both migratory and sedentary populations) exceeded those of sedentary lineages. Models of trait evolution indicated that partial migration is an intermediate step between migratory and sedentary states in this family. Given that partial migration is widespread across migratory animals (e.g., of all migratory birds, ca. 51% are partially migratory [5]), speciation via switches in migratory behavior might be an important yet overlooked mechanism of animal diversification.

Genetic structure in hybrids and progenitors provides insight into processes underlying an invasive cattail (Typha × glauca) hybrid zone.

Traditional models of hybrid zones have assumed relatively low hybrid fitness, and thus focussed more on interspecific gene flow than on hybrid dispersal. Therefore, when hybrids have high fitness and the potential for autonomous dispersal, we have limited understanding of whether hybrid dispersal or repeated local hybrid formation is more important for maintaining hybrid zones. The invasive hybrid cattail Typha × glauca occupies an extensive hybrid zone in northeastern North America where it is sympatric with its progenitors T. latifolia and T. angustifolia. We characterized genetic diversity and genetic structure of the three taxa across a broad spatial scale where the maternal parent is relatively rare, and tested the hypothesis that the hybrid shows stronger evidence of gene flow than its progenitor species, particularly among disturbed sites (ditches) compared with established wetlands. Support for this hypothesis would suggest that dispersal, rather than repeated local formation, is more important for maintaining hybrid zones. Within each taxon, genetic differentiation among ditches was comparable to that among wetlands, although clonal richness was consistently greater in ditches, suggesting more frequent seed establishment. Genetic structure across sites was more pronounced in the hybrid compared with either progenitor species. Overall, our data reflect relatively low gene flow in hybrids, and suggest that hybrids are more likely to be created in situ than to be introduced from other sites. Despite the high fitness of invasive T. × glauca and its potential for autonomy, local processes appear more important than dispersal in maintaining this hybrid zone.

Habitat-based isolating barriers are not strong in the speciation of ecologically divergent squirrels (Tamiasciurus douglasii and T. hudsonicus)

Adaptations to different environments between closely related species can be important drivers of reproductive isolation during speciation due to either habitat isolation between species or reduced fitness in hybrids that possess suboptimal adaptations. Hybrid zones are useful natural arenas to explore how ecologically divergent species compete for habitat in sympatry and how possible differences in their habitat use may contribute to the speciation process. We investigated habitat selection by sister species of tree squirrels, the Douglas squirrel and the red squirrel, that have evolved in different forest types in allopatry and hybridize in a transitional forest. We first used genome-wide SNP data and admixture analyses to classify individuals into parental or hybrid classes. Next, we estimated home ranges with radio telemetry data and then used a novel ground-based lidar system to measure forest canopy structure of squirrel home ranges, midden sites, and marginally used forest habitat. We found hybrids consisting of multiple hybrid classes were intermixed with both parental species in the same forest with varying canopy structure complexity. On average, Douglas squirrels utilized forests with slightly greater structural complexity than either red squirrels or hybrids, while marginally used forests were the least structurally complex. Interestingly, hybrid squirrels were not relegated to marginal habitat and were successful in mating among each other and with both parental species. As such, our study suggests that prezygotic-ecological isolation and postzygotic-hybrid infertility, and postzygotic ecological inviability of hybrids are not strong barriers in the speciation process between Douglas squirrels and red squirrels. Closely related species that become geographically divided often encounter different environments and thus evolve different adaptations. Sometimes, these species meet again and produce hybrids. Hybrids often have lower fitness due to their inferior adaptations. We studied a pair of closely related squirrels that evolved in different environments and meet and produce hybrids in a secondary contact zone. Despite these species evolving in allopatry in very different forest types, they do not select different forest characteristics while in sympatry in the hybrid zone. Furthermore, hybrids do not show major differences in types of habitats that they choose and are capable of defending territories and reproducing among each other and with both parental species. In summary, prezygotic and postzygotic isolating mechanisms associated with habitat selection do not appear to have an important role in promoting species divergence.

A simultaneous fit of vt = 0 and 1 torsion-wagging-rotational levels of CH3NH2 using a hybrid “tunneling and non-tunneling” Hamiltonian formalism

About 15,000 published transitions for methyl amine with J ≤ 40 have been refit using a hybrid program in which the NH2 wagging motion is treated by a tunneling formalism, while the CH3 internal-rotation motion is treated by the usual torsion-rotation Hamiltonian. Most of the fitted transitions are taken from published measurements. Compared to previous fits using the high barrier tunneling formalism for both motions, the present hybrid fit reduces the number of parameters in the Hamiltonian by about a factor of two, with essentially no increase in observed-minus-calculated residuals for the infrared transitions, and with an increase of about a factor of two for the microwave transitions. We showed that our approach can treat simultaneously for the first time the ground and first excited states of methylamine and can now be applied to (i) a much larger set of accurate vt = 1 microwave data, and (ii) infrared data on the vt = 2 ← 1 and 3 ← 2 torsional hot bands.

A Genetic Programming-Driven Data Fitting Method

Data fitting is the process of constructing a curve, or a set of mathematical functions, that has the best fit to a series of data points. Different with constructing a fitting model from same type of function, such as the polynomial model, we notice that a hybrid fitting model with multiple types of function may have a better fitting result. Moreover, this also shows better interpretability. However, a perfect smooth hybrid fitting model depends on a reasonable combination of multiple functions and a set of effective parameters. That is a high-dimensional multi-objective optimization problem. This paper proposes a novel data fitting model construction approach. In this approach, the model is expressed by an improved tree coding expression and constructed through an evolution search process driven by the genetic programming. In order to verify the validity of generated hybrid fitting model, 6 prediction problems are chosen for experiment studies. The experimental results show that the proposed method is superior to 7 typical methods in terms of the prediction accuracy and interpretability.

Fitness of F1 hybrids between stacked transgenic rice T1c-19 with cry1C*/bar genes and weedy rice

Compared to single-trait transgenic crops, stacked transgenic plants may be more prone to become weedy, and transgene flow from stacked transgenic plants to weedy relatives may pose a potential environmental risk because these hybrids could be more advantageous under specific environmental conditions. Evaluation of the potential environmental risk caused by stacked transgenes is essential for assessing the environmental consequences caused by crop-weed transgene flow. The agronomic performance of fitness-related traits was assessed in F1+ (transgene positive) hybrids (using the transgenic line T1c-19 as the paternal parent) in monoculture and mixed planting under presence or absence glufosinate pressure in the presence or absence of natural insect pressure and then compared with the performance of F1– (transgene negative) hybrids (using the non-transgenic line Minghui 63 (MH63) as the paternal parent) and their weedy rice counterparts. The results demonstrated that compared with the F1– hybrids and weedy rice counterparts, the F1+ hybrid presented higher performance (P<0.05) or non-significant changes (P>0.05) under natural insect pressure, respectively, lower performance (P<0.05) or non-significant changes (P>0.05) in the absence of insect pressure in monoculture planting, respectively. And compared to weedy rice counterparts, the F1+ hybrid presented higher performance (P<0.05) or non-significant changes (P>0.05) in the presence or absence of insect pressure in mixed planting, respectively. The F1+ hybrids presented non-significant changes (P>0.05) under the presence or absence glufosinate pressure under insect or non-insect pressure in monoculture planting. The all F1+ hybrids and two of three F1– hybrids had significantly lower (P<0.05) seed shattering than the weedy rice counterparts. The potential risk of gene flow from T1c-19 to weedy rice should be prevented due to the greater fitness advantage of F1 hybrids in the majority of cases.

Recombining Your Way Out of Trouble: The Genetic Architecture of Hybrid Fitness under Environmental Stress.

Hybridization between species can either promote or impede adaptation. But we know very little about the genetic basis of hybrid fitness, especially in nondomesticated organisms, and when populations are facing environmental stress. We made genetically variable F2 hybrid populations from two divergent Saccharomyces yeast species. We exposed populations to ten toxins and sequenced the most resilient hybrids on low coverage using ddRADseq to investigate four aspects of their genomes: 1) hybridity, 2) interspecific heterozygosity, 3) epistasis (positive or negative associations between nonhomologous chromosomes), and 4) ploidy. We used linear mixed-effect models and simulations to measure to which extent hybrid genome composition was contingent on the environment. Genomes grown in different environments varied in every aspect of hybridness measured, revealing strong genotype–environment interactions. We also found selection against heterozygosity or directional selection for one of the parental alleles, with larger fitness of genomes carrying more homozygous allelic combinations in an otherwise hybrid genomic background. In addition, individual chromosomes and chromosomal interactions showed significant species biases and pervasive aneuploidies. Against our expectations, we observed multiple beneficial, opposite-species chromosome associations, confirmed by epistasis- and selection-free computer simulations, which is surprising given the large divergence of parental genomes (∼15%). Together, these results suggest that successful, stress-resilient hybrid genomes can be assembled from the best features of both parents without paying high costs of negative epistasis. This illustrates the importance of measuring genetic trait architecture in an environmental context when determining the evolutionary potential of genetically diverse hybrid populations.

Gene Flow from Transgenic PPO-inhibiting Herbicide-resistant Rice to Weedy Rice, and Agronomic Performance by Their Hybrids

Hybridization of transgenic crops with their weedy relatives may spread transgenes into weedy populations and alter the ecological fitness of such hybrids. Our study objective was to quantify the rate of gene flow from herbicide-resistant transgenic rice to weedy rice and compare the performance of hybrids between them over several generations. The transgenic plants used in these experiments are resistant to protoporphyrinogen IX oxidase (PPO)-inhibiting herbicides such as oxyfluorfen, oxadiazon, and butafenacil. The rate of gene flow from those plants to five weedy accessions ranged from 0.025% to 0.139%. The F1 hybrids that resulted from gene flow from transgenic to weedy rice and subsequent F2 progeny were shorter than the weedy parents but produced significantly more and heavier grains. The F3 progeny resulting from gene flow were resistant to oxyfluorfen, which indicated that the gene associated with PPO-inhibiting herbicide resistance is dominantly inherited. The GR50 values (dose required for 50% shoot fresh weight reduction) for F3 progeny were > 69-fold greater than for the weedy rice. These results suggest that transgenes that could escape from PPO-inhibiting herbicide-resistant rice and be transferred to weedy rice might persist and disperse into weedy populations over several generations due to herbicide resistance and improved reproductive traits in the hybrids.

Phenotypic selection under two contrasting environments in wild sunflower and its crop-wild hybrid.

Hybridization is a common phenomenon in plants and can lead to the introgression of alleles from one population into another, generate new hybrid lineages, or cause species extinction. The environmental conditions and the genetic background of the participating populations may influence these outcomes since they can affect the fitness of hybrids, thereby increasing or decreasing the chances of introgression. Thus, it is important to understand the context‐dependent prospects for introgression of alleles into diverse populations and under multiple ecological environments. Crop–wild hybridization presents an opportunity to explore these dynamics in agroecosystems. To this end, we used diverse wild and hybrid sunflowers from across the northern United States as a basis for evaluating variation in morphological traits and assessing context‐dependent selection. These crop–wild hybrids and their wild counterparts were grown under agricultural conditions in the field with and without wheat competition. Interactions between origin and cross type affected expression of early functional traits, while interactions between competition and cross type acted on reproductive traits. A smattering of early and reproductive traits was affected by interactions between cross type and competition that varied by origin (i.e., 3‐way interactions). Seven functional traits, especially number of branches and tertiary head diameter, underwent net and direct directional selection, while six out of these seven traits appear to also be experiencing nonlinear selection dynamics. In general, wild‐like traits were favored under both sets of conditions, while, under wheat competition, some crop‐like traits related to fast growth and primary head diameter became important. These data reaffirm the hypothesis that stressful conditions establish a scenario more suitable for crop introgression and clarify that nonlinear selection dynamics may play a role in this process.

Limited intrinsic postzygotic reproductive isolation despite chromosomal rearrangements between closely related sympatric species of small ermine moths (Lepidoptera: Yponomeutidae)

AbstractIn evolutionarily young species and sympatric host races of phytophagous insects, postzygotic incompatibility is often not yet fully developed, but reduced fitness of hybrids is thought to facilitate further divergence. However, empirical evidence supporting this hypothesis is limited. To assess the role of reduced hybrid fitness, we studied meiosis and fertility in hybrids of two closely related small ermine moths, Yponomeuta padella and Yponomeuta cagnagella, and determined the extent of intrinsic postzygotic reproductive isolation. We found extensive rearrangements between the karyotypes of the two species and irregularities in meiotic chromosome pairing in their hybrids. The fertility of reciprocal F1 and, surprisingly, also of backcrosses with both parental species was not significantly decreased compared with intraspecific offspring. The results indicate that intrinsic postzygotic reproductive isolation between these closely related species is limited. We conclude that the observed chromosomal rearrangements are probably not the result of an accumulation of postzygotic incompatibilities preventing hybridization. Alternative explanations, such as adaptation to new host plants, are discussed.

Admixture mapping in interspecific Populus hybrids identifies classes of genomic architectures for phytochemical, morphological and growth traits.

Summary The genomic architecture of functionally important traits is key to understanding the maintenance of reproductive barriers and trait differences when divergent populations or species hybridize. We conducted a genome‐wide association study (GWAS) to study trait architecture in natural hybrids of two ecologically divergent Populus species.We genotyped 472 seedlings from a natural hybrid zone of Populus alba and Populus tremula for genome‐wide markers from reduced representation sequencing, phenotyped the plants in common gardens for 46 phytochemical (phenylpropanoid), morphological and growth traits, and used a Bayesian polygenic model for mapping.We detected three classes of genomic architectures: traits with finite, detectable associations of genetic loci with phenotypic variation in addition to highly polygenic heritability; traits with indications for polygenic heritability only; and traits with no detectable heritability. For the first class, we identified genome regions with plausible candidate genes for phenylpropanoid biosynthesis or its regulation, including MYB transcription factors and glycosyl transferases.GWAS in natural, recombinant hybrids represent a promising step towards resolving the genomic architecture of phenotypic traits in long‐lived species. This facilitates the fine‐mapping and subsequent functional characterization of genes and networks causing differences in hybrid performance and fitness.