Statistical and population genetic analyses
All statistical analysis was performed in R v. 3.6.1 (R Core Team,
2019). Genetic diversity (π ) (Nei, 1987) and Tajima’s Dtest statistic (Tajima, 1989) were estimated in the package pegas(Paradis, 2010). Analysis of molecular variance (AMOVA) (Excoffier,
Smouse, & Quattro, 1992) was performed using the package poppr(Kamvar, Brooks, & Grünwald, 2015) whereas FST(Nei, 1987) andDXY (R. R.
Hudson, Slatkin, & Maddison, 1992) were estimated with the packagePopGenome (Pfeifer, Wittelsbürger, Ramos-Onsins, & Lercher,
2014).
Standard two-side, unpaired t-tests were run on plasmid number between
genospecies comparing the two geographic populations using the functiont.test from the base R package (R Core Team, 2019). Classical
multidimensional scaling (MDS) was run using the cmdscalefunction using the base R package on a distance matrix calculated from
the binary presence/absence plasmid data per isolate. Further effects on
plasmid content were tested using a generalized linear mixed effects
model assuming a Poisson error distribution using the glmerfunction from the package lme4 (Bates, Maechler, Bolker, &
Walker, 2015). Fixed effects were included for sample origin (Asia vs.
Europe) and source (human vs. tick isolate) and genospecies was fitted
as a random effect. Mean estimates and their 95% credible intervals
were estimated based on 5000 simulations using the sim function
from the package arm (Gelman & Su, 2016). Residual error was
calculated according to Nakagawa & Schielzeth (2010).