NSF awards UAF $289,888 to build software tracing species hybridization

Tracing how species interbreed over evolutionary time, a process called hybridization, complicates efforts to build accurate evolutionary trees, and a University of Alaska Fairbanks project aims to build statistical software that fixes that. The National Science Foundation awarded UAF $289,888 for the work. The award abstract notes such tools also help trace relationships among viral strains, including the one causing COVID-19, aiding understanding of disease spread and virulence. The grant runs Sept. 1, 2026, through Aug. 31, 2029; NSF dated the award action Aug. 20.

Standard phylogenetic tree models can fail to represent reticulate processes such as hybridization, where genomic material from several parent species combines in new offspring. The project develops estimators under the Network Multispecies Coalescent Model, inferring networks rather than trees from full genome sequences or thousands of gene trees. The resulting software will be freely available and validated against simulated and empirical data sets.

Principal investigator Elizabeth S. Allman, a UAF mathematics professor, studies the statistical soundness of tree and network construction methods. She coauthored a 2024 paper on identifiability of level-1 species networks from gene tree quartets.

The grant also funds a summer school and teaching materials. It falls under NSF's Mathematical Biology program in the Directorate for Mathematical and Physical Sciences, part of the agency's long-running support for computational phylogenetics.