Hybridization can restructure gene co-expression networks

Simulations reveal that the effects of hybridization can linger for several generations.

Understanding the relationship between genotype and phenotype remains one of the central challenges in biology. How does an organism’s genetic make-up give rise to its observable traits? One way to approach this question is by studying the regulatory networks that govern gene expression. By examining how genes interact, researchers can uncover the underlying patterns that shape biological function. These networks are often inferred from statistical correlations between the expression levels of different genes. In simple terms, we can expect the expression of transcription factors to be positively correlated with the genes they activate, and the expression of inhibitors to be negatively correlated with their targets.

However, correlation does not necessarily imply causation. Many patterns of correlated gene expression may arise from indirect or unrelated molecular processes. It is therefore important to interpret gene co-expression patterns with caution. This task becomes even more complex in the presence of hybridization, as highlighted by a recent simulation study published in the journal Genetics.

Two Scenarios

Rogini Runghen and Daniel Bolnick examined two scenarios using individual-based simulations. In the first, they modeled successive generations of hybrid matings derived from two divergent parental populations. In the second, they simulated the effect of recurrent gene flow from a source population into a larger population with distinct gene expression patterns.

The analyses revealed that “these evolutionary processes can create blocks of highly correlated gene expression that might be identified as modules in a gene expression network, but which may have no shared biological function.” In other words, hybridization can lead to statistical correlations between genes that have no functional or regulatory meaning.

Three-phase Trajectory

A closer examination of the simulations showed a three-phase trajectory. In the initial disruption phase, hybrids exhibit substantial network perturbation, marked by a dramatic increase in gene network connectivity during the F2 generation. This disruption is followed by a reorganization phase, during which recombination in subsequent generations progressively breaks down the physical linkage between blocks of correlated gene expression. Finally, the networks enter a stabilization phase, characterized by greater structural organization than in the parental genotypes. Interestingly, hybrid networks retained elevated connectivity (approximately two-fold higher) even after 20 generations.

It is relatively intuitive that hybridization can generate statistical correlations between genes located on the same chromosome through physical linkage. However, the impact of hybridization extends even further, influencing correlations among genes situated on different chromosomes.

Even the correlations between genes on different chromosomes become denser. This happens because upstream regulatory genes in the network are co-inherited with other genes on the same linkage block, so all these linked genes become correlated with the downstream targets of their neighboring regulatory gene. 

Overall, these results demonstrate that hybridization can profoundly restructure gene regulatory networks, generating lasting patterns of elevated connectivity that extend both within and across chromosomes.

Simulations show how correlations between genes (in the first generations of hybrids) gradually decay due to recombination. From: Runghen & Bolnick (2025).

Evolutionary Innovation

This study provides a clear demonstration of how hybridization can shape statistical patterns of gene co-expression. Given the prevalence of hybridization across the Tree of Life, we must thus be cautious when interpreting correlations between genes. At the same time, the findings highlight the potentially creative role of hybridization: some hybridization-induced correlations may represent genuine functional connections, providing hybrids with novel expression patterns that open pathways to new phenotypes. Incorporating selection into future simulations could shed light on how such patterns of gene expression are refined (or eliminated) over evolutionary time. Ultimately, hybridization may not only act as a source of disruption, but also as a powerful engine of evolutionary innovation in gene regulatory networks.

References

Runghen, R., & Bolnick, D. I. (2025). Effects of hybridization and gene flow on gene co-expression networks. Genetics, 230(2), iyaf057.

Featured image: DNA strands © geralt | Wikimedia Commons

Can hybridization explain the evolution of plumage patterns in black-and-white wagtails?

A reticulated phylogeny provides extra challenges to answer this question.

Time flies when you are having fun. Several years ago, I wrote a blog post about the evolution of plumage patterns in the White Wagtail (Motacilla alba). The work of Georgy Semenov and his colleagues suggested that a small toolkit of genes might have been shuffled around by hybridization, resulting in the different head patterns of several subspecies.

These patterns are confined to a small number of patches – throat, back and sides of the head and neck – which can be either black or grey. Think shuffling a deck of cards and randomly extracting a combination of cards: black throat, grey back, black on the sides.

While this intriguing hypothesis has yet to be formally tested in the White Wagtail, a recent study in Systematic Biology examined this idea on a broader scale. Loïc Rancilhac and his colleagues investigated whether hybridization might account for the repeated emergence of similar plumage patterns across different wagtail species.

An Entangled Mess

The researchers focused on five species of “black-and-white wagtails”, namely the White Wagtail (M. alba), the African Pied Wagtail (M. aguimp), the Japanese Wagtail (M. grandis), the White-browed Wagtail (M. maderaspatensis), and the Mekong Wagtail (M. samveasnae). To explore parallel evolution of plumage patterns, we need a solid phylogenetic framework. But that was easier said than done. Genomic analyses revealed extensive phylogenetic conflict, with all 15 possible topologies recovered. It was thus not straightforward to determine the species tree (i.e. the topology that represents the actual evolutionary history).

The analyses were further complicated by signs of ancient introgression events. This additional difficulty is nicely illustrated by the uncertain phylogenetic position of the African Pied Wagtail (M. aguimp), which might involve introgression from an extinct or unsampled ghost lineage (a cool phenomenon that I have reviewed in this paper).

Thus, we consider two likely scenarios regarding the position of M. aguimp: either it is sister to M. alba and was introgressed by an unsampled lineage branching between the ingroup and M. cinerea, or it is sister to the other four black-and-white wagtail species and extensive ancient introgression occurred with M. alba.

Furthermore, allele sharing between two geographically separated species – the African Pied Wagtail and the Japanese Wagtail – suggests that a third species, most likely the White Wagtail, acted as a “genetic bridge” between them. I have outlined this general scenario in my review on multispecies hybridization, and a similar case has recently been documented in Darwin’s finches (see this blog post).

Two possible phylogenetic networks that could explain the genetic patterns in black-and-white wagtails. The black lines represent candidate species trees, while the dotted black branch in the right network is a hypothetical ghost lineage. Red arrows indicate introgression events. From: Rancilhac et al. (2024).

Plumage Puzzle

Despite this phylogenetic uncertainty, the researchers were still able to investigate the evolution of plumage patterns in these wagtails. They noted that “while introgressive hybridization has occurred, inferred reticulations do not connect lineages with similar phenotypes, contradicting the hypothesis that parallel plumage evolution is caused by horizontal transfer of alleles.” Instead, the repeated plumage patterns may result from the sorting of ancestral variation or lineage-specific de novo mutations. Future research will be needed to determine the relative contribution of new mutations, ancestral variants, and introgressive hybridization in the evolutionary history of these birds (see these blog posts on wheatears and Darwin’s finches for similar cases). For now, the puzzle of plumage patterns in black-and-white wagtails remains to be solved.

References

Alaei Kakhki, N., Schweizer, M., Lutgen, D., Bowie, R. C., Shirihai, H., Suh, A., Scheilzeth, H. & Burri, R. (2023). A phylogenomic assessment of processes underpinning convergent evolution in open-habitat chats. Molecular Biology and Evolution, 40(1), msac278.

Ottenburghs, J. (2019). Multispecies hybridization in birds. Avian Research, 10(1), 20.

Ottenburghs, J. (2020). Ghost introgression: spooky gene flow in the distant past. Bioessays, 42(6), 2000012.

Rancilhac, L., Enbody, E. D., Harris, R., Saitoh, T., Irestedt, M., Liu, Y., Lei, F., Andersson, L. & Alström, P. (2024). Introgression underlies phylogenetic uncertainty but not parallel plumage evolution in a recent songbird radiation. Systematic Biology, 73(1), 12-25.

Rubin, C. J., Enbody, E. D., Dobreva, M. P., Abzhanov, A., Davis, B. W., Lamichhaney, S., et al. (2022). Rapid adaptive radiation of Darwin’s finches depends on ancestral genetic modules. Science Advances, 8(27), eabm5982.

Semenov, G. A., Koblik, E. A., Red’kin, Y. A., & Badyaev, A. V. (2018). Extensive phenotypic diversification coexists with little genetic divergence and a lack of population structure in the White Wagtail subspecies complex (Motacilla alba). Journal of Evolutionary Biology, 31(8), 1093-1108.

Featured image: White Wagtail (Motacilla alba) © J.M. Garg | Wikimedia Commons

Something borrowed: Ancestral variation and introgression drive rapid evolution in Darwin’s Finches

Genomic study identifies 28 ancestral genetic modules that predate the radiation.

Despite the overwhelming evidence for evolution, many misconceptions still persist (often actively promoted by creationists). One example is the so-called “waiting time problem” which claims that populations must wait for the right mutation to appear before they can adapt to a new environment. This view grossly misrepresents how evolution actually works. In addition to novel mutations, populations can draw on ancestral genetic variation and even DNA introgressed from closely related species. What remains an open question, however, is the relative contribution of new mutations, ancestral variants, and introgressive hybridization to evolutionary change.

A recent study in Science Advances offers a compelling example of how ancestral variation and introgression have fueled the adaptive radiation of Darwin’s finches. The researchers found that multiple ancestral genetic modules have been exchanged among species, contributing to the remarkable phenotypic diversity that characterizes this iconic group of birds.

Twenty-eight Loci

Carl-Johan Rubin, Erik Enbody and their colleagues compared the genomes of three Geospiza species that differ in their beak size: Small Ground Finch (G. fuliginosa), Medium Ground Finch (G. fortis) and Large Ground Finch (G. magnirostris). Using admixture mapping, they identified 28 genomic regions significantly associated with variation in beak morphology. These regions were clustered on macrochromosomes and included the previously characterized ALX1 and HMGA2 genes, which are known to influence beak shape.

Next, the researchers explored the evolutionary origins of these 28 genomic regions by comparing their genetic make-up across the phylogeny of Darwin’s finches. This approach revealed that these regions predate the divergence of the genera Geospiza and Camarhynchus. Further analyses, along with previous studies, also provided evidence of introgression among several species.

Genome-wide admixture mapping uncovered 28 regions (highlighted in red) that are associated with beak morphology in three Geospiza species. From: Rubin et al. (2022).

Rapid Radiation

Taken together, these findings provide a clear illustration of how ancestral variation and introgression can fuel rapid adaptive radiations. The authors nicely summarize the main message at the end of the introduction.

We show that the origin of these haplotype blocks linked to phenotypic divergence predates speciation events. These genetic modules have been reused over the past million years, were exchanged by gene flow, and contributed to the rapid phenotypic evolution and speciation among Darwin’s finches.

There’s no need to wait for new mutations when evolution can work with what already exists.

References

Rubin, C. J., Enbody, E. D., Dobreva, M. P., Abzhanov, A., Davis, B. W., Lamichhaney, S., et al. (2022). Rapid adaptive radiation of Darwin’s finches depends on ancestral genetic modules. Science Advances, 8(27), eabm5982.

Featured image: Medium Ground Finch (Geospiza fortis) © Judy Gallagher | Wikimedia Commons

Three-way hybridization between flameback woodpeckers on Sri Lanka

A colorful combination of cryptic speciation and hybridization.

Running the Avian Hybrids blog has been a constant source of inspiration for my own research. To showcase this connection, I recently added a page listing scientific publications shaped by ideas and discussions from the website. These papers fall into two categories: broad overviews of avian hybridization and studies focused on particular bird groups. Let me quickly highlight two of my favorite papers.

In 2019, I wrote an extensive review on multispecies hybridization in birds. The main message is nicely captured in the first sentence of the abstract: “Hybridization is not always limited to two species; often multiple species are interbreeding.” Several years later, in 2024, I decided to compile an overview of hybridization in woodpeckers. In addition, I explored the evolutionary trade-off between hybridization and mimicry in this wonderful group of birds (see this blog post for the details).

These two themes – multispecies hybridization and woodpeckers – come together in a recent paper by Rashika Ranasinghe and her colleagues in the journal Ecology and Evolution. Let’s travel to Sri Lanka to see what is going on there.

Three Genetic Clusters

The island of Sri Lanka is home to two species of flameback woodpeckers: the endemic Red-backed Flameback (Dinopium psarodes) in the south and the Black-rumped Flameback (D. benghalense) in the north. Earlier studies have already shown that these species interbreed in the island’s north-central region, producing a striking gradient of plumage colors – from deep red through orange to golden yellow. But the new genomic analyses added an unexpected twist to the story: instead of just two species interbreeding, the data revealed three genetically distinct populations engaged in hybridization.

The woodpeckers fall neatly into three genetic clusters: (1) birds from Mannar Island on the northwestern coast, (2) birds from Jaffna Peninsula in the far north, and (3) birds from the rest of Sri Lanka. In the north-central contact zone, however, these boundaries start to blur. Individuals in this region carry genetic ancestry from all three groups. In other words: a three-way hybrid zone!

Admixture analyses uncovered three genetic clusters, corresponding to Mannar Island (blue), Jaffna Peninsula (green) and Sri Lanka (red). Individuals from the contact zone carry genetic ancestry from all three clusters. From: Ranasinghe et al. (2024).

Evolutionary History

How did this three-way hybrid zone arise? Based on phylogenetic analyses that included other Dinopium species, the researchers suggest a possible scenario. The flamebacks on Sri Lanka probably descended from a historical colonization event of D. b. tehminae from mainland India. Divergence times point to the late Pleistocene (126,000-11,700 years ago) when lowered sea levels exposed a land bridge connecting India and Sri Lanka. Once on the island, the woodpeckers gradually split into distinct populations, shaped by isolation-by-distance, genetic drift, and local adaptation.

However, it is also possible that ongoing gene flow has blurred the phylogenetic signal in these woodpeckers. In my own work on the Bean Goose complex, for example, we showed how introgression can influence branching patterns across large portions of the genome (see this blog post). To unravel the evolutionary history of the Red-backed and Black-rumped Flamebacks, more detailed analyses will be needed. Hence, plenty of burning questions about these flamebacks remain to be answered.

References

Ranasinghe, R. W., Seneviratne, S. S., & Irwin, D. (2024). Cryptic hybridization dynamics in a three‐way hybrid zone of Dinopium flamebacks on a tropical island. Ecology and Evolution, 14(12), e70716.

Featured image: Black-rumped Flameback (Dinopium benghalense) © Suyothami | Wikimedia Commons