Imputation of ancient canid genomes reveals inbreeding history over the past 10,000 years.

The multi-millenia long history between dogs and humans has justly placed them at the forefront of archeological and genomic research. Despite ongoing efforts including the analysis of ancient dog and wolf genomes, many questions remain regarding their geographic and temporal origins, and the microevolutionary processes that led to the huge diversity of breeds today. Although ancient genomes provide valuable information, their use is significantly hindered by low depth of coverage and post-mortem damage, which often inhibits confident genotype calling. In the present study, we assess how genotype imputation of ancient dog and wolf genomes, utilising a large reference panel, can improve the resolution afforded by ancient genomic datasets. Imputation accuracy was evaluated by down-sampling 10 high coverage ancient and modern dog and wolf genomes to 0.05-2x coverage and comparing concordance between imputed and high coverage genotypes. We also measured the impact of imputation on principal component analyses (PCA) and runs of homozygosity (ROH). Our findings show high (R 2 > 0.9) imputation accuracy for dogs with coverage as low as 0.5x and for wolves as low as 1.0x. We then imputed a worldwide dataset of 81 published ancient dog and wolf genomes, in addition to nine newly sequenced medieval and early modern period European dogs, to assess changes in inbreeding during the last 10,000 years of dog evolution. Ancient dog and wolf populations generally exhibited lower inbreeding levels than present-day individuals, though with some exceptions occurring in ancient Arctic and European dogs. Interestingly, regions with low ROH density maintained across ancient and present-day samples were significantly associated with genes related to olfaction and immune response. Our study indicates that imputing ancient canine genomes is a viable strategy that allows for the use of analytical methods previously limited to high-quality genetic data.