We used 20 de novo genome assemblies to probe the speciation history and architecture of gene flow in rapidly radiating Heliconius butterflies. Our tests to distinguish incomplete lineage sorting from introgression indicate that gene flow has obscured several ancient phylogenetic relationships in this group over large swathes of the genome. Introgressed loci are underrepresented in low-recombination and gene-rich regions, consistent with the purging of foreign alleles more tightly linked to incompatibility loci. Here, we identify a hitherto unknown inversion that traps a color pattern switch locus. We infer that this inversion was transferred between lineages by introgression and is convergent with a similar rearrangement in another part of the genus. These multiple de novo genome sequences enable improved understanding of the importance of introgression and selective processes in adaptive radiation.

We here pioneer a low-cost assembly strategy for 20 Heliconiini genomes to characterize the evolutionary history of the rapidly radiating genus Heliconius. A bifurcating tree provides a poor fit to the data, and we therefore explore a reticulate phylogeny for Heliconius. We probe the genomic architecture of gene flow, and develop a new method to distinguish incomplete lineage sorting from introgression. We find that most loci with non-canonical histories arose through introgression, and are strongly underrepresented in regions of low recombination and high gene density. This is expected if introgressed alleles are more likely to be purged in such regions due to tighter linkage with incompatibility loci. Finally, we identify a hitherto unrecognized inversion, and show it is a convergent structural rearrangement that captures a known color pattern switch locus within the genus. Our multi-genome assembly approach enables an improved understanding of adaptive radiation.

Chromosomal inversions are ubiquitous in genomes and often coordinate complex phenotypes, such as the covariation of behavior and morphology in many birds, fishes, insects or mammals [1][2][3][4][5][6][7][8][9][10][11] . However, why and how inversions become associated with polymorphic traits remains obscure. Here we show that despite a strong selective advantage when they form, inversions accumulate recessive deleterious mutations that generate frequency-dependent selection and promote their maintenance at intermediate frequency. Combining genomics and in vivo fitness analyses in a model butterfly for wing-pattern polymorphism, Heliconius numata, we reveal that three ecologically advantageous inversions have built up a heavy mutational load from the sequential accumulation of deleterious mutations and transposable elements. Inversions associate with sharply reduced viability when homozygous, which prevent them from replacing ancestral chromosome arrangements. Our results suggest that other complex polymorphisms, rather than representing adaptations to competing ecological optima, could evolve because chromosomal rearrangements are intrinsically prone to carrying recessive harmful mutations.Many organisms display concerted variation in their phenotypic traits. Consistent association of multiple phenotypic features, combining differences in behavior, morphology and physiology, may result in so-called syndromes, or complex traits with clear adaptive significance. This coordination is often controlled by chromosomal rearrangements. Examples include dimorphic social organization in several ant species 7 , color displays and mating behaviors in many birds and butterflies [3][4][5][6]12 , dimorphic flower morphology in plants 13 , as well as the extreme cases provided by sexual dimorphism in numerous animals. Why and how these structurally-delimited complex polymorphisms arise is a long-standing puzzle in biology 11,14-17 . The so-called supergenes controlling the coordination of the multiple phenotypic features are characterized by suppression of recombination, often through polymorphic chromosomal rearrangements, which preserve alternative combinations of alleles at linked genes 4,7,11,13 . The encoded multi-feature phenotypes are often assumed to reflect the existence of multiple, distinct adaptive optima, and their maintenance in polymorphisms to result from antagonistic ecological factors such as differential survival or mating success 8,12,18,19 . Yet why and how polymorphic