Asymmetrical hybridization and environmental factors influence the spatial genetic structure of a killifish hybrid zone

Hybridization offers insight into speciation and the forces that maintain barriers to reproduction, and hybrid zones provide excellent opportunities to test how environment shapes barriers to reproduction and hybrid fitness. A hybrid zone between the killifish, Fundulus heteroclitus and F. grandis, had been identified in northeastern Florida, although the spatial structure and parameters that affect the distribution of the two species remain unknown. The present study aimed to determine the fine-scale spatial genetic patterns of the hybrid zone to test the hypothesis that species ranges are influenced by changes in dominant vegetation, and to determine how differences in reproductive barriers between the two species influence the observed patterns. The area of overlap between the two species spanned ~37 km and showed a mosaic pattern of hybridization, suggesting the spatial structure of the hybrid zone is largely influenced by the environment. Environmental association analysis, however, suggested that while dominant vegetation had a significant influence on the spatial structure of the hybrid zone, a combination of environmental factors was driving the observed patterns. Hybridization tended to be rare at sites where F. heteroclitus was the more abundant species, suggesting that differences in preference for conspecifics can lead to differences in rates of introgression into parental taxa and likely result in a range-shift as opposed to adaptation in the face of climate change.

Methods
DNA was extracted from fin tip tissue of 431 total individuals of Fundulus heteroclitus, Fundulus grandis, and Fundulus hybrids. Library was prepared according to Peterson et al. 2012 (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0037135) for ddRADseq using restriction endonucleases SphI-HF and MluCI and a size selection window of 375-450 base pairs. Individuals were sequenced in three separate lanes at the University of Florida Interdisciplinary Center for Biotechnology Research on an Illumina NovaSeq 6000 (S4, 2x100). Raw sequence reads were imported into the STACKS pipeline (Rochette et al. 2019; https://onlinelibrary.wiley.com/doi/abs/10.1111/mec.15253), PCR duplicates were removed, and sequences were demultiplexed. The data present here include all resulting demultiplexed sequences used in the project.