Format: ORAL

David L. Field1, Sean Stankowski2, Arka Pal3, Parvathy Surendranadh3, Desmond Bradley4, Enrico Coen4, Nick Barton3

1 Macquarie University, Sydney, Australia 2 University of Sussex, UK 3 Institute of Science and Technology, Klosterneuburg, Austria, 4 John Innes Centre, Norwich, UK

A fundamental question in evolutionary biology is how phenotypes and their underlying genes interact to generate reproductive barriers between populations. The fitness landscape has been an influential framework to model these connections, yet direct measures of how multiple genes interact to create fitness valleys between populations remains challenging to obtain in nature. I will present some of the work weve been doing to address this gap by exploiting a natural hybrid zone between subspecies of Antirrhinum majus (snapdragons) with a well-defined genotype-phenotype map. This work includes investigations into the molecular genetic basis of flower colour, ecological experiments on the agents of selection, and the development of novel population genomic approaches to detect barriers in the genome. A key focus will be on our multi-generational study, where-by genotyping and phenotyping over 30,000 plants over 14 years, weve obtained a pedigree-based estimate of the fitness landscape separating these recently diverged subspecies. At deeper evolutionary time scales, using whole-genomes along geographic clines we find the absence of a genome-wide barrier to gene flow with only a small proportion of the genome resisting introgression. By linking recent ecological time-scales of fitness variation in nature to patterns of deeper genomic divergence, our goal is to provide a general model for how reproductive barriers arise during the early stages of divergence.