Ion Torrent data for the genome assembly and phylogenomic placement of mitochondrial genomes with a focus on houndsharks (Chondrichthyes: Triakidae)

Here, we present the Ion Torrent® next-generation sequencing (NGS) data for five houndsharks (Chondrichthyes: Triakidae), which include Galeorhinus galeus (17,487 bp; GenBank accession number ON652874), Mustelus asterias (16,708; ON652873), Mustelus mosis (16,755; ON075077), Mustelus palumbes (16,708; ON075076), and Triakis megalopterus (16,746 bp; ON075075). All assembled mitogenomes encode 13 protein-coding genes (PCGs), two ribosomal (r)RNA genes, and 22 transfer (t)RNA genes (tRNALeu and tRNASer are duplicated), except for G. galeus which contains 23 tRNA genes where tRNAThr is duplicated. We also present our code and corresponding datasets used to assemble and annotate their mitogenomes, prepare alignments, partition our datasets, assign models of evolution, infer phylogenies based on traditional site homogeneous concatenation approaches as well as under the multispecies coalescent model (MSCM) and site heterogenous models, and generate statistical data for comparison of different topological outcomes. The data and code presented in this paper can assist other researchers in further elucidating the diversification of triakid species and the phylogenetic relationships within Carcharhiniformes (groundsharks) as mitogenomes accumulate in public repositories.

Methods
Data collection
Genomic DNA extraction: Standard CTAB protocol or SDS-based lysis buffer (PL2) from the NucleoSpin Plant II mini kit (MACHEREY-NAGEL, Dueren, Germany); DNA quality control: Qubit 4.0 fluorometer (ThermoFisher Scientific) and LabChip® GXII Touch (PerkinElmer, Waltham, MA, USA); Library preparation: Ion Plus Fragment Library Kit (ThermoFisher Scientific) according to the manufacturer’s protocol, Ion Xpress™ Plus gDNA Fragment Library Preparation User Guide (MAN0009847 K.0); Polymerase chain reaction: 50 ng template DNA (Galeorhinus galeus genomic DNA), 1X GoTaq Buffer, 2.5 mM MgCl2, 200 µM dNTPs, 0.3 µM of each primer [Cytb CC F (5’- ACTTGAATTGGAGGGCAACC-3’) and Dloop Gga R (5’- AGGGTATGTGGGCCATATCA -3’)] and 0.625 U GoTaq DNA polymerase in a total reaction volume of 15 µL using the SimpliAmp™ Thermo Cycler machine. Cycling parameters: (i) one initial denaturation cycle at 94°C for 3 min, (ii) denaturation at 94°C for 30 s, annealing for 30 s starting at 65°C and decreasing in 1°C increments to 55°C for 11 cycles and then maintaining 55°C for a further 24 cycles and elongation at 72°C for 1 min, and (iii) a final elongation cycle at 72°C for 10 min; PCR quality control : 1.5% agarose gel electrophoresis at 100 Volts for 1 hour and visualisation using the Gel Doc™ XR+ documentation system (Bio- Rad); NGS Sequencing: Ion GeneStudio S5 Prime System and postprocessing with Torrent Suite version 5.16 under default settings to generate BAM reads and cycle sequencing with standard Sanger sequencing chemistry (BigDye® Terminator v.3.1 cycle sequencing kit) and capillary electrophoresis on the ABI3730xl genetic analyser (Applied Biosystems®) for FASTA reads. Conducted at the Central Analytical Facility (CAF) at Stellenbosch University. 

Data processing
For the five houndshark species for which sequencing data was generated, sequence quality was checked in FastQC, adaptors and poor-quality bases (Phred score below 20) were trimmed, and reads shorter than 25 bp were removed in Torrent Suite Version 5.16. Raw reads were aligned to the Mustelus mustelus mitogenome (NC_039629.1) using the Geneious read mapper with medium sensitivity settings and five iterations in Geneious Prime v.2023.2 (Data 1-5). The reads that mapped to the reference mitogenome were then saved in BAM format as filtered Ion Torrent reads to use for mitogenome assembly in SPAdes v.3.15 with the input set for unpaired Ion Torrent reads with 8 threads, kmers 21,33,55,77,99,127, the careful option to reduce the number of mismatches and short indels and all other parameters left as default. A third assembly approach was conducted by mapping the original raw reads de novo in SPAdes using the same parameters. Sanger sequence reads were generated for the Galeorhinus galeus mitogenome to confirm a structural deviation (tandem duplication and random loss mutation) in the mitogenome and trimmed using Finch TV v.1.5 (Data  6-7). All three assemblies were compared to produce a consensus sequence for each species and these were annotated with MitoAnnotator in MitoFish v.3.72. Mitogenome sequence files in GenBank format were generated and uploaded onto NCBI (ON075075, ON075076, ON075077, ON652873, and ON652874) (Data 8-12).

The five newly assembled mitogenomes were used in an extensive statistical workflow to expand the phylogeny of Triakidae. First, we downloaded additional representative mitogenomes from each family in the order Carcharhiniformes and four representative species each from the orders Lamniformes and Orectolobiformes from GenBank to complete the Galeomorphii cluster (Data 13) and produced codon-aware multiple sequence alignments (MSA) for each of the 13 PCGs in MACSE v.2.07 and the 2 rRNAs in MAFFT v.7.299. We concatenated the PCG alignments (Data 14-16), concatenated the PCG and rRNA alignments (Data 17-19), and translated and then concatenated the PCG alignments (Data 20-22). These concatenated alignments were used for phylogenetic reconstruction under Maximum Likelihood (ML) and Bayesian Inference (BI) approaches in IQTree v.2.1.3 and MrBayes v.3.2.6 as well as multispecies coalescent modelling in ASTRAL v.5.6.3 and SVDQuartets in PAUP* v4.0a 169 (a modified input NEXUS file with gene partitions is available as Data 23). Site heterogenous models were applied in PHYLOBAYES_MPI v.1.9. We designated eight different a priori partitioning schemes with varying degrees of complexity to determine the effects of partitioning on phylogeny estimation (PS01-PS08) for ML and BI analyses.