STAR: Theory and Software for Detecting Speciation with Periodic Gene Flow
Ente: Evo Patterns & Processes
Scadenza: 2029-09-30
Importo max: 399.733 EUR
Paese: US
Descrizione
New species often form when populations become geographically isolated and undergo independent evolutionary change. When natural selection for different traits is strong enough to overcome the blending effect of interbreeding, species can also form without geographic isolation. Scientists typically view these pathways to speciation as strict alternatives, but the geologic record suggests a mix of the two may be more realistic for many organisms. This project will develop mathematical models to understand how quickly speciation occurs when periods of isolation are interspersed with periods of interbreeding. It will also develop software that uses artificial intelligence to infer speciation history from modern DNA sequences. This tool will be used to study a group of birds that inhabit islands that have intermittently connected by land bridges during periods of low sea level, providing a strong test of the method. The project will advance biology by producing new theory and software that may be useful for diverse fields that use genomes to study the history of populations. It will benefit education by training a postdoctoral researcher and graduate students in modern quantitative methods and by supporting a workshop for STEM high school teachers in Montana.
Common models of speciation with gene flow consider constant migration or admixture after an
initial period of allopatry, but Earth’s recent climatic history suggests that many populations have
experienced cycles of isolation and contact. This project will use mathematical models and methods development to assess how periodic gene flow during divergence impacts speciation dynamics and whether its signature can be detected with genomic data. Specifically, it will develop theoretical expectations of time to speciation under different migration and natural selection regimes, validate model predictions with evolutionary simulations, and develop demographic inference methods using machine learning tools. These methods will be tested through application to a pair of young Melanesian bird lineages originally studied by Ernst Mayr. Because these taxa have limited dispersal ability and are found on islands that were intermittently connected by land bridges over the past million years, they likely experienced periodic gene flow during divergence. As “pulse” models of admixture are widely applied in contemporary population genomics, the project’s results may also inform understanding of complex demographic histories below the species level. Machine learning is increasingly applied in population genetics, but remains rare in speciation research; the project’s open source software may stimulate more widespread use. Specimens and empirical data resulting from this study will improve understanding of the historical connectivity and population size of species in a threatened biodiversity hotspot.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using
Istituzione: Montana State University
Sede: BOZEMAN, MT
PI: Ethan Linck
Settori: Biological Sciences
Vai al bando originale
Registrati gratis su Bandolo per trovare bandi compatibili con la tua azienda.