Mitogenomes: genetic diversity, population structure, and demographic inference
There were 36 unique mitochondrial haplotypes in the dataset that included close relatives (n = 83), and 34 among the 58 “unrelated” individuals. All subsequent analyses were performed with the unrelated dataset. H d was 0.977 (SD 0.008), and π was 0.00583 (SD 0.0006). Phylogenetic analyses recovered mountain treeshrews as a monophyletic group with two deeply divergent but sympatric lineages. The average number of nucleotide substitutions per site between the two lineages is 0.013 (Figure S2). Outgroup relationships were consistent with the topology in Roberts et al. (2011). The BEAST dating analysis suggests that the lineages diverged ca. 450,000 ybp (95% Highest Posterior Density, HPD, 346,000–631,900 ybp) (Figure S3).
The median joining haplotype network (Figure 3) shows that the two mountain treeshrew haplogroups are sympatric on both MT and MK. Three haplotypes are found on both mountains (Table S3). Including related individuals, haplogroups 1 and 2 are found in near equal proportion on MK and MT (16 and 14 individuals, respectively), while haplogroup 1 is more frequent on MT (46 out of 53 individuals) (Figure 4a). The AMOVA on the unrelated dataset showed significant differentiation between the two mountains (F ST = 0.133, p =.00812), with 13.3% of variance accounted for by differences between mountains and 86.7% within mountains. Dividing the population into high (≥ 2000 masl) and low (< 2000 masl) elevation groups on each peak, 90.42% of the total variance is accounted for by within-group variation, and 9.58% among (F ST = 0.096, p = 0.027). Pairwise comparisons showed significant differences between high elevation MK and low elevation MT (F ST = 0.15,p = 0.023) and high elevation MK and high elevation MT (F ST = 0.18, p = 0.013).
Tajima’s D test was not significant, indicating a lack of evidence for recent population contraction or expansion. In the Bayesian skyline plot analysis, the 95% HPD of the population change parameter included zero, meaning that we cannot reject the hypothesis of zero demographic changes in the last 60,000 years.