Abstract
Tasmanian devils (Sarcophilus harrisii) are evolving in response to a unique transmissible cancer, devil facial tumour disease (DFTD), first described in 1996. Persistence of wild populations and the recent emergence of a second independently evolved transmissible cancer suggest that transmissible cancers may be a recurrent feature in devils. Here, we compared signatures of selection across temporal scales to determine whether genes or gene pathways under contemporary selection (six to eight generations) have also been subject to historical selection (65–85 Myr). First, we used targeted sequencing, RAD-capture, in approximately 2500 devils in six populations to identify genomic regions subject to rapid evolution. We documented genome-wide contemporary evolution, including 186 candidate genes related to cell cycling and immune response. Then we used a molecular evolution approach to identify historical positive selection in devils compared to other marsupials and found evidence of selection in 1773 genes. However, we found limited overlap across time scales, with only 16 shared candidate genes, and no overlap in enriched functional gene sets. Our results are consistent with a novel, multi-locus evolutionary response of devils to DFTD. Our results can inform conservation by identifying high priority targets for genetic monitoring and guiding maintenance of adaptive potential in managed populations.
Highlights
Species are subject to selection by pathogens throughout their evolutionary history, shaping lineage diversification and leading to complex cellular and molecular defensive mechanisms [1]
Chromosomal fragility [39], a reportedly high incidence of non-transmissible neoplasms [40] and injury-prone biting behaviour [41] may contribute to a predisposition to transmissible cancers in devils [42]. These findings suggest that transmissible cancers may be a recurring selective force in the Tasmanian devil lineage
We identified candidate single-nucleotide polymorphisms (SNPs) as the top 1% of a de-correlated composite of multiple signals (DCMS) score [50], which combined the results of three analyses: change in allele frequency in each population after devil facial tumour disease (DFTD) (Δaf ), and two methods that estimate strength of selection from allele frequencies at multiple time points in multiple populations, the method of Mathieson & McVean [14], which allows the estimated selection coefficient to vary over space; and spatpg [15], which allows the selection coefficient to vary over time and space
Summary
Species are subject to selection by pathogens throughout their evolutionary history, shaping lineage diversification and leading to complex cellular and molecular defensive mechanisms [1]. Intra-specific comparisons across populations can reveal to what extent the evolutionary response to disease is constrained by limited genetic mechanisms or variation for adaptation [16] Reduced representation techniques such as restriction-site associated DNA-sequencing (RADseq) [17] have made the acquisition of genome-wide, time-series genetic data more accessible in non-model systems [18]. By integrating these resources and tests of selection at differing temporal scales, we can assess whether species that show rapid evolution in response to contemporary pathogens show evidence of historical selection to similar pathogens.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
