Genotyping-by-sequencing of 731 leuciscid minnows across 25 stream sites in southern Ontario, Canada, found hybrids at every site (29.4% of individuals), spanning 29 of 45 possible species-pair crosses, plus 40 multi-species hybrids and two unidentifiable 'cryptic' ancestry components (ENTROPY clusters 11 and 12). The study could not resolve what causes this hybridization: watershed-scale land-use predictors showed only weak, outlier-driven associations; the mechanisms (species abundance, shared breeding behaviour, or habitat alteration) remain unidentified; whether some species act as 'hub' species that transmit ancestry between otherwise non-hybridizing species is unknown; the sources of the two cryptic ancestry clusters (candidate ghost admixture from unsampled populations or species) could not be determined; and it is unresolved whether the gene flow is primarily historical (post-glacial) or recent and disturbance-driven. This task is to determine the drivers, pathways, and temporal dynamics of this documented multi-species hybridization: identify the mechanisms and environmental predictors that explain variation in hybridization frequency, establish the network structure of introgression (hybrids of some species backcrossing with other species, hub species, versus direct multi-way crosses), identify the sources of the cryptic ancestry clusters, resolve whether gene flow is historical or contemporary, and settle the admixture patterns of central stoneroller, whose estimates carry wide credible intervals.

Background

Leuciscidae is the family of freshwater fishes that includes most North American 'minnows' and 'shiners'; its species often co-occur in small streams, are morphologically hard to tell apart, and have weak reproductive isolation, so natural hybridization has been recorded in the group since at least Hubbs (1955). Classic allozyme and mitochondrial work showed that hybridization between the common shiner (Luxilus cornutus) and striped shiner (Luxilus chrysocephalus) has produced introgression — gene flow between species through hybrid backcrossing — extending well beyond their contact zone and replacing the mitochondrial DNA of striped shiner populations (Dowling and Hoeh 1991). Across North American minnows generally, breeding behaviour (for example nest association, where species spawn over other species' nests) statistically predicts which species hybridize in nature (Corush et al. 2020), and a community-level genomic survey of Ozark freshwater fishes found hybrids concentrated in Leuciscidae, at about half the sampled communities, with hybrid occurrence predictable from a few environmental variables (Zbinden et al. 2023). Meuser, Pitura, McFarlane and Mandeville (2025, published 2026 in Molecular Ecology) brought this to the community scale in southern Ontario. Using genotyping-by-sequencing (a reduced-representation method that genotypes thousands of single nucleotide polymorphisms genome-wide) on 731 individuals of nine target species — creek chub (Semotilus atromaculatus), common shiner, central stoneroller (Campostoma anomalum), hornyhead chub (Nocomis biguttatus), river chub (Nocomis micropogon), western blacknose dace (Rhinichthys obtusus), longnose dace (Rhinichthys cataractae), striped shiner, and rosyface shiner (Notropis rubellus) — plus ENTROPY (a Bayesian Markov-chain-Monte-Carlo program that estimates each individual's ancestry proportions from genotype data) and DNA barcoding, they found hybrids at all 25 sites; 29.4% of individuals had at least 10% heterospecific ancestry; 29 of 45 possible species-pair crosses were detected; two previously unrecorded parental species (fathead and bluntnose minnow, Pimephales spp.) appeared as untargeted ancestry sources; and 40 individuals (5.5%) were multi-species hybrids carrying ancestry from three or more sources. A best-fitting model with twelve ancestry clusters (K = 12, selected by the deviance information criterion) required two clusters (11 and 12) that matched no sampled species: each contributed no more than 1.2% of any individual's ancestry yet the model with them fit better, which the authors interpret as possible 'ghost admixture' — ancestry from unsampled populations or species, whose distorting effect on inferences from incomplete sampling was analyzed by Beerli (2004). Logistic regressions of per-site hybrid proportion on watershed-scale land-use variables from the Ontario Watershed Information Tool were all non-significant and outlier-driven, and the study explicitly could not identify the mechanisms controlling hybridization outcomes. The conceptual frame for the network question comes from multispecies-hybridization theory: when more than two species interbreed, the interactions form a network in which some taxa may be 'hub' species that hybridize with many others and potentially transmit ancestry between species that do not hybridize directly (Ottenburghs 2019); analogous three-species systems show that such network structure can be reconstructed from genomic data (Natola et al. 2022). Whether backcrossing of hybrids with other species, hub-mediated ancestry transmission, or direct multi-way mating produces multi-species hybrids is unknown in general and in this assemblage in particular. The timing question has deep context: common shiner × striped shiner hybridization has been ongoing at least since the Pleistocene (Dowling and Hoeh 1991), and serial mitochondrial-genome replacement among Luxilus species driven by Pleistocene glacial cycles has been demonstrated in the Ozarks (Duvernell et al. 2025), but the origin of the community-wide gene flow documented in southern Ontario — continuous since post-glacial recolonization, or accelerated by recent anthropogenic disturbance — was not determined and could not be from these data alone. All raw data, scripts, and metadata for the study are public (NCBI Sequence Read Archive BioProject PRJNA1291971; Dryad doi:10.5061/dryad.5mkkwh7kn), so the open questions are answerable by reanalysis combined with new sampling.

Problem Statement

For the nine-species leuciscid minnow assemblage of southern Ontario streams documented by Meuser et al. (2026), determine the drivers, pathways, and temporal dynamics of its extensive multi-species hybridization, jointly resolving: (1) the sources of the two cryptic ancestry components (ENTROPY clusters 11 and 12) — identifying them as specific unsampled populations or species, or otherwise establishing what they represent; (2) the network structure of introgression — whether multi-species hybrids arise because hybrids of some species are backcrossing with other species, because particular species act as 'hub' species that transmit ancestry between otherwise non-hybridizing species, or through direct multi-way crosses; (3) the timing of gene flow — whether the hybridization is primarily historical, ongoing since post-glacial recolonization, or a recent phenomenon accelerated by anthropogenic disturbance; (4) the mechanisms and environmental predictors that explain variation in hybridization frequency across sites — testing the roles of species abundance, shared breeding behaviours, and habitat alteration, measured at the spatial scale (watershed-level land use versus finer riparian or local habitat characteristics) that best explains hybridization; and (5) the admixture patterns involving central stoneroller (Campostoma anomalum), whose current estimates carry wide credible intervals. These components together explain the causes, pathways, and origin of the documented hybridization; an answer that resolves a proper subset should state explicitly which components it settles and how the remaining ones are constrained, but must in every case resolve at least one of components (2)–(4) — the network, timing, or driver components that carry the stated objective — and may leave any of the others as constraints.

References

  1. Meuser, A. V., A. R. Pitura, S. E. McFarlane, and E. G. Mandeville. 2025. 'Extensive multi-species hybridization between Leuciscidae minnow species.' bioRxiv 2025.07.14.664671. doi:10.1101/2025.07.14.664671. https://doi.org/10.1101/2025.07.14.664671
  2. Meuser, A. V., A. R. Pitura, S. E. McFarlane, and E. G. Mandeville. 2026. 'Extensive Multi-Species Hybridization Between Leuciscidae Minnow Species.' Molecular Ecology 35(6): e70301. doi:10.1111/mec.70301. https://doi.org/10.1111/mec.70301 (published version, PMC13013738)
  3. Zbinden, Z. D., M. R. Douglas, T. K. Chafin, and M. E. Douglas. 2023. 'A community genomics approach to natural hybridization.' Proceedings of the Royal Society B 290: 20230768. doi:10.1098/rspb.2023.0768. https://doi.org/10.1098/rspb.2023.0768
  4. Ottenburghs, J. 2019. 'Multispecies hybridization in birds.' Avian Research 10: 20. doi:10.1186/s40657-019-0159-4. https://doi.org/10.1186/s40657-019-0159-4
  5. Duvernell, D. D., C. Arnold, S. Koju, A. J. Wicks, and T. E. Dowling. 2025. 'A Complicated History of Mitogenome Introgression Among Luxilus Species (Teleostei, Family Leuciscidae) in the Ozark Highlands.' Fishes 10(9): 443. doi:10.3390/fishes10090443. https://doi.org/10.3390/fishes10090443
  6. Corush, J. B., B. M. Fitzpatrick, E. L. Wolfe, and B. P. Keck. 2020. 'Breeding behaviour predicts patterns of natural hybridization in North American minnows (Cyprinidae).' Journal of Evolutionary Biology 34: 486-500. doi:10.1111/jeb.13751. https://doi.org/10.1111/jeb.13751
  7. Beerli, P. 2004. 'Effect of unsampled populations on the estimation of population sizes and migration rates between sampled populations.' Molecular Ecology 13: 827-836. doi:10.1111/j.1365-294x.2004.02101.x. https://doi.org/10.1111/j.1365-294x.2004.02101.x
  8. Dowling, T. E., and W. R. Hoeh. 1991. 'The Extent of Introgression Outside the Contact Zone Between Notropis cornutus and Notropis chrysocephalus (Teleostei: Cyprinidae).' Evolution 45: 944-956. doi:10.1111/j.1558-5646.1991.tb04362.x. https://doi.org/10.1111/j.1558-5646.1991.tb04362.x
  9. Hubbs, C. L. 1955. 'Hybridization between Fish Species in Nature.' Systematic Zoology 4: 1-20. doi:10.2307/2411933. https://doi.org/10.2307/2411933
  10. Natola, L., S. S. Seneviratne, and D. E. Irwin. 2022. 'Population genomics of an emergent tri-species hybrid zone.' Molecular Ecology 31: 5356-5367. doi:10.1111/mec.16650. https://doi.org/10.1111/mec.16650