Results
For phylogenetic and population genetics analyses, we focused on the
linear chromosome as it is a core genomic compartment present in allBorrelia and is generally used to reconstruct the evolutionary
history between genospecies (Becker et al., 2020; S. R. Casjens et al.,
2012; Mongodin et al., 2013). Borrelia genomes are highly
fragmented and can contain over 20 unique linear and circular plasmids
(S. R. Casjens et al., 2012; Fraser et al., 1997) which can be highly
plastic even within a single genospecies (Becker et al., 2020; S. R.
Casjens et al., 2018; Mongodin et al., 2013). Borrelia plasmids
contain genes related to host and vector adaptation and absence of
certain plasmid types have been linked to reduced infectivity (S. R.
Casjens et al., 2018; Dulebohn, Bestor, & Rosa, 2013; Ellis et al.,
2013; Embers, Alvarez, Ooms, & Philipp, 2008; Grimm et al., 2004, 2005;
Lin, Diuk-Wasser, Stevenson, & Kraiczy, 2020). Therefore, their
presence could influence the evolution of these bacteria and should be
considered. As each plasmid carries specific, partitioning genes
(categorized to PFam32, 49, 50, 57/62) and generally exists in single
copies per cell (S. Casjens & Huang, 1993; S. R. Casjens et al., 2012;
Fraser et al., 1997), we were able to approximate plasmid content in
each isolate by searching for these partitioning genes using BLAST
v.2.8.1 (Altschul et al., 1990; Camacho et al., 2009) (algorithm:blastn )
In total 142 full chromosome sequences were used for further population
genetics analysis, of which 136 were assembled de novo from
Illumina MiSeq data. Final chromosome length ranged from 825.7 to 906.2
kb (mean: 900.0kb) with chromosome coverages from 8 to 572x (mean:
104.6x) (see Table S1). Plasmid content could only be estimated for the
136 samples for which raw MiSeq data was available. For full isolate
information see Table S1.