Inferring ancestral population type from limited samples

Introduction A population’s demographic “type” is characterized by the shape of its age distribution. This structure provides insights into its behavioral ecology, such as how elderly individuals contribute to the survival and reproductive success of younger members (e.g., the grandmother hypothesis). As human populations vary across a broad spectrum of types—ranging from enslaved (Type II) to modern industrial (Type I) populations—there is significant interest in identifying the demographic shapes of ancient populations. Methods We simulated allele frequency drift for 17 age-structured populations with overlapping generations across a range of population types. Differences between estimates of age-specific loss of genetic diversity were utilized to estimate survivorship curves ( l x ), which were then used to categorize the source population’s type. Results We demonstrate that Pollak’s F k — a measure designed to capture changes in allele frequencies across time points within an age-class — can be accurately approximated across different age-classes within a single time point (i.e., a standing population). Discussion In scenarios where DNA samples are limited but span multiple age-classes, such as in ancient DNA or endangered species research, this method makes it possible to estimate population type and demographic history.