Highlight: Ghosts of Long-Lost Connections in the Genomes of Wild Rice Authors: Pedro Andrade Published: 2026 Charles Darwin's iconic sketch of a bifurcating tree representing species’ evolutionary history has influenced biologists for generations. Increasingly, though, molecular evolutionists are uncovering footprints of extensive reticulate evolutionary histories whereby different genomic regions within a species can trace to different ancestral lineages (Mallet et al. 2016). Introgression, i.e. the transfer of genetic material between closely related lineages following hybridization, is a frequent explanation for these patterns among plants and animals, and the growing evidence of genetic exchange in natural systems has reframed our appreciation of its role in species formation. Rice, which includes species of the genus Oryza, is well known for its complex evolutionary history. The domesticated rice species, Oryza sativa and the less common Oryza glaberrima, belong to the so-called “AA-genome group” of Oryza, along with six wild rice species (Chen et al. 2019). It has generally been accepted that the species tree of the AA-group can be characterized by a bifurcating pattern with species clustering by geography (Fig. 1): Asian rice lineages are sister to the African group; their immediate outgroup is the American Oryza glumaepatula. The two other AA-genome species, the African Oryza longistaminata and the Australian Oryza meridionalis, are earlier diverging clades of AA-genome rice. Phylogenetic tree of Oryza AA-genome species, with indication of their main current geographical distribution. A gray bidirectional arrow indicates novel putative introgression events identified by Koyanagi et al. (2026). There is evidence that the true picture of rice evolution is less straightforward. For example, gene flow among domesticated and wild lineages of Asian rice suggests that these likely constitute a hybrid swarm and that domestic O. sativa is non-monophyletic (Wang et al. 2017). The analysis of endogenous viral elements in the genomes of AA-group rice species further supports non-bifurcating evolution (Saito et al. 2023). “Our recent analysis of virus-derived sequences in rice genomes had hinted at an unexpectedly close relationship between African O. longistaminata and Asian rice, two lineages that the standard branching tree places far apart,” explains Kanako Koyanagi, Associate Professor at Hokkaido University in Japan (Fig. 2). According to her, “this made us suspect that the branching tree was not the whole story.” GBE author spotlight: Kanako O. Koyanagi. Kanako O. Koyanagi is an associate professor in the Faculty of Information Science and Technology at Hokkaido University. She earned her PhD from Kyoto University, where she studied gene duplication in animals. She has long been interested in reading the past from present-day genomic and other molecular information using phylogenetic and comparative approaches. Her work spans a range of systems: for example, she has applied phylogenetic analysis to genome-wide epigenetic data to infer how cell types differentiate, as reported in Genome Biology and Evolution in 2015 (Koyanagi 2015). Her current focus is on rice. In Koyanagi et al. (2026), she searched the genomes of wild Oryza species for footprints of ancient introgression between lineages now separated by entire continents. She is now interested in the genomic changes that arose during the domestication and improvement of rice and other crops. In a new article now published in Genome Biology and Evolution, Koyanagi et al. (2026) surveyed the genomes of AA-group rice species for signatures of ancient gene flow between species. The authors hypothesized that phylogenetic incongruence between the consensus genome-wide phylogenetic tree and trees constructed from different genomic regions would reveal potential instances of between-species genetic exchange. An initial phylogenetic analysis based on a whole-genome alignment recovered a pattern consistent with previous findings (Fig. 1). Koyanagi et al. (2026) then analyzed 15,460 independent genome regions to test whether these regions showed the same pattern as the consensus phylogeny. Frequently, they did not. In 13.5% of the genomic regions analyzed by Koyanagi et al. (2026), O. meridionalis clustered with the Asian or American group; for O. longistaminata, 18.6% of the genome clustered with either the Asian, African, or American species. These discordant regions were located across the genome, meaning not one single region of the genome explains these non-consensus patterns. Such signatures of phylogenetic discordance can result from introgression, but also from incomplete lineage sorting, i.e. when ancestral genetic variation is differentially sorted among diverging daughter lineages. To discern between these two hypotheses, Koyanagi et al. (2026) performed ABBA-BABA analyses to test for introgression between each species of the Asian-African-American group and either O. meridionalis or O. longistaminata. The results support a scenario of introgression: in particular, Asian wild rice species share an excessive number of derived alleles with both O. meridionalis and O. longistaminata. Further supporting the hypothesis of introgression, the authors calculated divergence times between these groups and found that in the discordant genomic regions the median divergence times were more than 3-fold lower than the genome-wide divergence times. These findings suggest recent genetic exchange after species divergence. Oryza longistaminata currently does not occur in Asia. What could explain these patterns of genetic exchange? Koyanagi hypothesizes that “after the common ancestor of these lineages split, their geographical distributions overlapped for some time, and hybridization occurred during that period of overlap.” Alternatively, “another scenario would be that there was, or is, an unknown Asian species related to African O. longistaminata, which hybridized with Asian species,” Koyanagi adds. For the moment, these hypotheses remain to be tested. It also remains unclear whether introgressed alleles conferred adaptive benefits. Koyanagi has high hopes for this line of research, hoping that “future studies will reveal why these introgressed segments from ancestral Oryza species have persisted, and whether they have any functional significance.” For now, the authors have set their sights on obtaining new evidence from currently unsampled diversity; for example, obtaining genomes from field-collected accessions of O. longistaminata would fill an existing gap in the available wild rice data. Want to learn more? Check out these other articles on the role of introgression in plant evolution recently published in Genome Biology and Evolution: Blanco-Pastor JL. Alternative modes of introgression-mediated selection shaped crop adaptation to novel climates. Genome Biol Evol. 2022 Aug:14(8):evac107. https://doi.org/10.1093/gbe/evac107. Blanco-Pastor JL. Alternative modes of introgression-mediated selection shaped crop adaptation to novel climates. Genome Biol Evol. 2022 Aug:14(8):evac107. https://doi.org/10.1093/gbe/evac107. Vi T, Vigouroux Y, Cubry P, Marraccini P, Phan HV, Khong GN, Poncet V. Genome-wide admixture mapping identifies wild ancestry-of-origin segments in cultivated Robusta coffee. Genome Biol Evol. 2023:15(5):evad065. https://doi.org/10.1093/gbe/evad065. Vi T, Vigouroux Y, Cubry P, Marraccini P, Phan HV, Khong GN, Poncet V. Genome-wide admixture mapping identifies wild ancestry-of-origin segments in cultivated Robusta coffee. Genome Biol Evol. 2023:15(5):evad065. https://doi.org/10.1093/gbe/evad065.