Cristy Gelling is Communications Director at the GSA, a science writer, and a lapsed yeast geneticist.
Heliconius melpomene. See Davey et al. 6: 695–708. Image courtesy of Chris Jiggins. Processed using Textify (

The dizzying array of wing patterns in Heliconius butterflies has served as a model for evolution and adaptation in the wild for more than a century. The genus is most famous for the way different species within a geographic region tend to converge on similar wing markings—known as biological mimicry.

In the latest issue of G3, Davey et al. present a substantially improved assembly of the Heliconius melpomene genome created using methods that should be useful for other wild organisms. The new data confirmed that ten of the 21 Heliconius chromosomes are fusions of smaller chromosomes from the neighbouring Eueides genus, which may partly explain the high speciation rate within this diverse genus.

The first H. melpomene genome was generated from short-read sequencing and published in 2012. To match up the fragments of partly-assembled sequence (known as scaffolds) to specific chromosomes, the data was compared to a linkage map. These maps plot the position of genetic landmarks across the genome by analyzing the tendency of variants at neighboring genes to be co-inherited more frequently. However, this still left many gaps in the sequence and many cases where the orientation, order, or location of a chunk of sequence remained uncertain.

Since publication of the first version of the genome, sequencing methods that generate longer fragments have become much more effective. To create the improved assembly, Davey et al. used PacBio sequencing, a technology that generates longer reads—in the thousands, rather than hundreds of base pairs—to fill in some of the gaps in the original sequence. They also used a much more detailed linkage map created by sequencing a family of butterflies: the grandmother, two parents, and 69 offspring. The first linkage map was produced using Restriction-site associated DNA (RAD) sequencing, which placed markers roughly 10 kilobases apart; the new linkage map used whole genome sequencing, so recombination breakpoints could be identified to within hundreds of bases. This allowed much more accurate placing of scaffolds and identification of mis-assemblies. Finally, they improved the assembly by combining overlapping scaffolds that had been separated due to heterozygous variation (i.e., because the sequence on one chromosome was slightly different than the sequence of the matching, homologous chromosome).

The result was a much more complete assembly, with considerably fewer gaps and more sequence “anchored” to a specific location and orientation. For example, only 27% of the genome was anchored in the first genome assembly, compared to 84% in the improved version.

The new data allowed the team to test the hypothesis that the evolution of the Heliconius genome involved fusion of many chromosomes over a relatively short period. They were able to confirm that ten of the 21 Heliconius chromosomes are the result of fusion events that have occurred since the split from its sister lineages approximately 6 million years ago.

The authors speculate that this might explain why the speciation rate in Heliconius is much higher than in a closely related genus in which the chromosomes have remained separate. Longer chromosomes typically have lower recombination rates, i.e., the genome is less thoroughly “shuffled” with each generation. A variety of population genetics models predict that lower recombination rates promote the accumulation of genetic differences of a type that can lead to the formation of new species. Testing this hypothesis will reveal whether some of the diversity of this colorful genus can be traced back to its genome structure.



Davey, J. W., Chouteau, M., Barker, S. L., Maroja, L., Baxter, S. W., Simpson, F., Merrill, R. M., Joron, M., Mallet, J., Dasmahapatra, K. K.  & Jiggins, C. D. (2016). Major improvements to the Heliconius melpomene genome assembly used to confirm 10 chromosome fusion events in 6 million years of butterfly evolution. G3: Genes|Genomes|Genetics, 6: 695-708 doi: 10.1534/g3.115.023655

    Leave a comment