Figure 6
From: Inferring population size changes with sequence and SNP data: lessons from human bottlenecks
Effect of different models of population structure and different sampling strategies on Tajima’s D. Average values of Tajima’s D over 100 000 replicate simulations of divergence models with global sampling (left panel), island models with local sampling (central panel) and island models with global sampling (right panel), as function of the number of demes. The simulated samples are composed of 100 chromosomes (of length 50 000 bp), with a mutation rate of 1.2 × 10−8 per site and per generation and a recombination rate of 1.5 × 10−8 per site and per generation. In all models, the demes are of size 10 000 present-day diploid individuals, and n represents the number of demes. In the local sampling scheme, all chromosomes are sampled from one deme only. In the global sampling schemes, however, the chromosomes are sampled from all the demes, with an almost equal number of chromosomes within each deme (plus or minus one chromosome). No migration is allowed in the divergence model, whereas a migration rate of 1 × 10−3 migrants/generation is assumed in the island model.
