DarT data were produced based on samples gathered at IRD, IFREMER, and CSIRO with the partners associated, and genotyped using DarT protocols and genotyped by CSIRO using DarT protocols.

The data to download contain: 1/demultiplexed raw data, 2/data filtered using radiator and exported with different format (genpop, structure etc...), and 3/demographic simulations using FastSimCoal and GADMA.

A total of 364 individual DNAs passed the DArT library constructions’ quality checks and genotypes were generated from sequencing runs completed at DArT using a proprietary DArTseq analytical pipeline (DArT-Soft14 version). The DArT toolbox was then used to filter away poor-quality sequences and apply more stringent selection criteria to the barcode region, and generate the final genotypes (Kilian et al. 2012), here called “Raw data”. As detailed in the publication associated to these data (Nikolic et al. 2022, submitted), the data were transformed iteratively (Figure 1) following a sequence of filtering steps using radiator (Gosselin 2018; Gosselin et al. 2020) (see detailed in the publication associated). This resulted in a dataset of SNPs (one SNP per de novo assembled fragment) called “DATA 3” excluding sex-linked marked yet keeping outlier loci. The “DATA 4” was subsequently obtained by removing SNPs detected as outliers and those with minor allele frequencies (MAF) lower than 0.01 using dartR v1.5.5 (Georges et al. 2018). Finally, a dataset “DATA 5” including only the 9 outlier loci detected was also constructed for comparative population genetic analysis. When grouping individuals per geographic regions, the final number of individuals is 312 while a clustering per sampling site led to retain 313 individuals (Figure 1, see also the R Markdown file “2024_04_Blue_shark_markdown.pdf”). The dataset in this web portal contains also the data form to run demographic inferences through GADMA (Noskova et al. 2020) and FastSimCoal (Excoffier et al. 2011) analysis detailed in the associated publication.