Locked within our genetic code are the histories of our genes and the genes of our ancestors. Deciphering a population's history from genetic data often involves lengthy investigations of many loci for many individuals. We test hypothetical population histories of the Thule expansion using a new coalescent simulation method that uses little more than mitochondrial haplogroup data. This new methodology rejects a severe bottleneck at expansion and reveals the range of probable population histories on which to focus future research.
Models of genetic population structure generally assume that emigrants from each local group are drawn at random from the set of individuals born there. We show that small violations of this assumption can have disproportionately large effects on genetic population structure, and we introduce a statistical method for measuring this effect.
We estimate the strength of kin-structured migration in six human populations (five from New Guinea and one from Finland) and in one population of nonhuman primates. We also test the hypothesis that migration is not kin structured by generating a sampling distribution of the estimator under the null hypothesis of independent random migration. We are unable to detect a statistically significant level of kin-structured migration in any population. However, five of our six human populations were from Papua New Guinea, and we cannot dismiss the possibility that migration is kin structured in other parts of the world.