3.1 Genetic data collection, bioinformatic processing, and locus
assembly
Our concatenated c-mos and cyt b data set (Sanger data set
hereafter) consisted of 1237 bp, including indels. Both genes were
represented in all samples with the exception of c-mos for the outgroupContia longicaudae . After filtering (see Methods, above), our
genus-level ddRAD data set consisted of 2848 loci with 20.7% missing
data (here defined as proportion of missing loci across all
individuals), and an effective mean per-sample depth of coverage of
78.7x ± 13.6x. Our T. blandingii data set consisted of 7231 loci
with 5.0% missing data, and an effective mean per-sample depth of
coverage of 83.6x ± 12.0x. Our T. pulverulenta data set consisted
of 4471 loci with 7.3% missing data, and an effective per-sample mean
depth of coverage of 77.9x ± 14.6x.