TABLE 3.
Serotype | Isolates | Core SNPs (Pairwise Range)b | Clock/Population Modelc | Mean/Median Tree Height in Years (95% HPD Interval)d | Mean/Median Evolutionary Rate in Substitutions/Site/Year (95% HPD Interval)e |
Anatum | 15 | 337 (0–257) | Strict/Skyline | 1484.9/1837.0 (549.6–1980.1) |
1.67 × 10–7/1.48 × 10–7 (6.92 × 10–11–3.86 × 10–7) |
Cerro | 13 | 21 (0–12) | Strict/Skyline | 2008.2/2008.4 (2007.6–2008.6) |
9.11 × 10–7/8.94 × 10–7 (3.06 × 10–7–1.57 × 10–6) |
Kentucky | 36 | 102 (0–30) | Relaxed/Skyline | 2004.1/2005.0 (2000.8–2006.8) |
6.39 × 10–7/6.34 × 10–7 (2.05 × 10–7–1.07 × 10–6) |
Meleagridis | 19 | 27 (0–9) | Strict/Skyline | 2007.4/2007.5 (2006.9–2007.7) |
6.88 × 10–7/6.66 × 10–7 (2.90 × 10–7–1.12 × 10–6) |
Newport | 16 | 52 (0–38) | Relaxed/Skyline | 2004.2/2004.6 (2000.4–2007.8) |
9.02 × 10–7/8.22 × 10–7 (2.64 × 10–7–1.65 × 10–6) |
Typhimurium (Copenhagen) | 27 | 732 (0–634) | Relaxed/Skyline | 1936.0/1943.0 (1864.7–1991.4) |
1.07 × 10–6/9.66 × 10–7 (2.84 × 10–7–2.05 × 10–6) |
aSee Supplementary Table 5 for an extended version of this table; note that evolutionary rates may be higher than previously reported estimates for Salmonella populations isolated over a longer time frame, due to the small sample sizes and short temporal period characterized here (Moller et al., 2018).
bNumber of core SNPs identified among all genomes within the serotype after removing recombination with Gubbins; the range of pairwise SNP differences between isolates was calculated using the dist.gene function in the ape package in R.
cThe optimal model selected for the data set; can be a combination of a strict or lognormal relaxed molecular clock (“Strict” or “Relaxed,” respectively) and a Constant Coalescent or Coalescent Bayesian Skyline population model (“CC” or “Skyline,” respectively); see Supplementary Table 5 for more details.
dThe tree height parameter and its respective 95% highest posterior density (HPD) interval reported by Tracer.
eCorresponds to the clock Rate and rate.mean parameters estimated by BEAST2 for models using strict and lognormal relaxed molecular clock models, respectively, as reported by Tracer.