Mitochondrial genome skimming
Iterative rounds of genome skimming on 48 specimens allowed us to reconstruct the entire maternally-inherited mitogenome for Coelaturini from the Malawi Basin. This mitogenome contains 15,664 bp, and annotation included the 37 expected genes (Boore, 1999): 2 for rRNA, 22 for tRNAs, which encode components in the mitochondrial translational machinery, and 13 other genes that encode protein components of the respiratory chain and ATP synthase (Fig. 4). On average ~40% of the mitogenome was recovered per individual (range 0-100%), with consistent recovery of several gene regions across most individuals, which would enable integration of mitogenomic data in phylogenetic and population genetic analyses.
Macroevolutionary analyses
All 95 Coelaturini specimens were included in our phylogenetic analyses. Selection of loci with >50% taxon-completeness resulted in a dataset that contained 1,109 ORF supercontigs with a mean and total alignment length of 2,118 bp and 2,348,614 bp, respectively. For UCEs, the cleaned and trimmed dataset contained 276 loci with a mean and total alignment length of 432 bp and 119,105 bp, respectively. For these, 515,219 (21.9%) and 11,001 (9.24%) sites, respectively, were phylogenetically informative. Additional data on specimen representation, alignment length, and the proportion of informative and missing sites for each ORF and UCE locus are given in Ortiz-Sepulveda et al. (2022). The partition analysis resulted in a 402–partition scheme with 45 unique substitution models for ORFs, and a 155-partition scheme and 51 unique substitution models for UCEs (Ortiz-Sepulveda et al., 2022). The phylogenetic trees reconstructed from the ORF and UCE datasets separately are highly congruent (Fig. 5, S3, S4). Whereas the ORF tree is fully supported, except for at population-level branches within the ‘Malawi’ clade, some uncertainty exists as to terminal branches in the UCE tree, as evidenced by decreasing support and some branch re-arrangements at very shallow levels of divergence (Fig. 5, S3, S4).