2.6 | Estimates of the effective population size and
divergence time
The pairwise sequentially Markovian coalescent (PSMC) method (Heng Li &
Durbin, 2011) was used to evaluate the dynamic change of effective
population size (Ne ) of each population. We used 0.17 years per
generation (g ) and mutation rate (μ ) of
9×10-9 per generation per site to rescale the time to
year (Cutter, 2008). More recent (within 1000 years) changes of
effective population size of each population and separation time between
different populations were further estimated by using the MSMC2
(Schiffels & Durbin, 2014), which can much compensate results from
PSMC. However, the inference accuracy of MSMC2 largely depends on the
phasing accuracy of genotypes. Switch error rates will introduce bias in
the calculation. To further confirm results from multiple sequentially
Markovian coalescent (MSMC2), we also used Sequentially Markovian
Coalescent (SMC++) methods (Terhorst, Kamm, & Song, 2017) to do the
same analysis as of MSMC2. The SMC++ used phasing-free genotype data to
do the population history and separation time inference, which becomes a
reliable method to support inferences from MSMC2. For SMC++, we set the
upper bound for the number of generations to 10,000 to estimate size
history, and calculate the lower bound based on a heuristic approach.
For MSMC2, we first phased all SNPs of each individual by using beagle
(v5.0) (Browning & Browning, 2007), then the calculation was performed
with the following parameters: -i 20 -t 6 -p ’10*1+15*2’. The mutation
rate (μ ) of Parascaris spp. for SMC++ and MSMC2 were used
the same values as for PSMC.