*3.8. Genome Sequencing and Multilocus Sequence Analysis of CB1-14*

A draft genome sequence of the isolate CB1-14 was obtained using 454 GS Junior (Roche Life Science, USA). A *de novo* assembly was performed using Newbler version 3.0 software Junior (Roche Life Science, USA). The genome sequence was assembled into 621 contigs with 14,322 bp of N50. The estimated genome size was 5.3 Mb. Gene prediction and automated genome annotation were

carried out using RAST v. 2.0 with default parameters [30]. Sequences of eight protein-coding genes (*ftsZ*, *gapA*, *gyrB*, *mreB*, *pyrH*, *recA*, *rpoA*, and *topA*) from twenty taxa were retrieved from the CB1-14 draft genome, and from the GenBank/DDBJ/EMBL databases. The MEGA program was used to concatenate, align, and reconstruct the ML, maximum parsimony (MP), and neighbor-joining (NJ) phylogenies with 1000 bootstrap replications. The best-fit model for protein evolution determined in the MEGA program was HKY+G [31]. Split decomposition analysis was performed using SplitsTree version 4.14.3 with a neighbor net drawing and a Jukes–Cantor correction [32,33].
