Genetics of randomly bred cats support the cradle of cat domestication being in the Near East

Abstract

Cat domestication likely initiated as a symbiotic relationship between wildcats (Felis silvestris subspecies) and the peoples of developing agrarian societies in the Fertile Crescent. As humans transitioned from hunter-gatherers to farmers ~12,000 years ago, bold wildcats likely capitalized on increased prey density (i.e., rodents). Humans benefited from the cats' predation on these vermin. To refine the site(s) of cat domestication, over 1000 random-bred cats of primarily Eurasian descent were genotyped for single-nucleotide variants and short tandem repeats. The overall cat population structure suggested a single worldwide population with significant isolation by the distance of peripheral subpopulations. The cat population heterozygosity decreased as genetic distance from the proposed cat progenitor's (F.s. lybica) natural habitat increased. Domestic cat origins are focused in the eastern Mediterranean Basin, spreading to nearby islands, and southernly via the Levantine coast into the Nile Valley. Cat population diversity supports the migration patterns of humans and other symbiotic species.

Fig. 1. Principal component analyses (PCA) of genetic variation in random-bred and wildcat felines.

a PCA plot of SNP data (N = 983). b PCA plot of STR data (N = 1861). A single point represents an individual, the shape represents a geographic region, the color represents a geographic sub-region. The two wildcat populations are denoted by squares of different colors. Middle Eastern, South Asia, and Western European cats form the peripheral subpopulations of random-bred cats. The wildcat hybrids and the island population of San Marcos, Baja California, are additional peripheral populations for a and b, respectively.

Fig. 2. Random-bred cat population SNP fastSTRUCTURE plot of K = 3.

Population contributions are represented by different colors, individual vertical bars represent an individual, and populations are separated by black lines.

Fig. 3. Random-bred cat population STR STRUCTURE plot of K = 3.

Population contributions are represented by different colors, individual vertical bars represent an individual, and populations are separated by black lines.

Fig. 4. Comparison of geographic distance and genetic distance of random-bred cat populations.

a Plot of SNP data with 24 sample locations with a regression line indicating a correlation of 0.447 with a p value of 0.001. b Plot of STR data with 22 sample locations with a regression line indicating a correlation of 0.302 with a p value of 0.0076. Each point represents an individual pairwise comparison of sample location populations.

Fig. 5. Comparison of genetic distance from F.s. lybica and observed heterozygosity.

a Plot of SNP data with a regression line indicating a correlation of −0.57 with a p value of 0.0034. b Plot of STR data with a regression line indicating a correlation of −0.33 with a p value of 0.13. Each point represents a sample location population, shape represents a geographical region, and color represents a geographical sub-region.

Fig. 6. Observed heterozygosity by sample location of SNP data for random-bred cat populations of Eurasia.

Each point represents a sample location population with the color showing the calculated observed heterozygosity. The triangle shape indicates an admixed population with a significant f₃ statistic, and the circle shape represents non-admixed populations. Populations of yellow and light orange shades are focused in the Near East and Mediterranean Basin.