In a world where habitats are degrading and the climate is warming and becoming unpredictable, biodiversity conservation efforts and funding remain grossly inadequate. As part of a multi-faceted approach to halting extinctions, shifting from preserving small, remnant populations to a practice of genetically connecting populations that recreate larger and more diverse populations is expected to be beneficial. This harnesses key evolutionary processes to promote species’ abilities to adapt to changing environments and to increase the likelihood of population persistence. We use the endangered Macquarie perch (Macquaria australasica) as a case study to develop a genetic strategy for metapopulation management aimed at promoting population growth and persistence. The Macquarie perch’s range has been highly fragmented, and its remaining habitat is at risk of catastrophic degradation due to climate change. We integrate results of new and existing genetic analyses to illustrate how genetically depauperate populations can benefit from admixture, and how the outcomes of management interventions can be quantified through genetic monitoring. We develop the pipeline JeDi (https://github.com/drobledoruiz/JeDi) for estimating unbiased individual heterozygosity, population nucleotide diversity,y and pairwise population divergence, using reduced-representation genome sequencing data and an assembled reference genome. We use this pipeline to estimate baseline data for monitoring of Macquarie perch populations and show that combining two genetic sources of migrants during population restoration resulted in doubling of nucleotide diversity compared to either source. Genetic diversity estimates using our pipeline are comparable across studies, datasets and species, and suitable for evaluating the rate of global biodiversity change.

To obtain data for variant and invariant sites for all individuals, we used raw sequencing data provided by DArT. Pre-processing of raw reads included trimming Illumina adaptors with fastp (Chen et al. 2018), and demultiplexing with process_radtags of STACKS package (Catchen et al. 2013). We mapped the processed reads to the chromosome-length Macquarie perch genome using BWA-MEM (Li and Durbin 2009) with parameter -R set to '@RG\tID:sample_name\tSM:sample_name'. Mapped reads per individual were sorted, filtered (min MAPQ ≥ 20) and transformed to BAM with samtools (Li et al. 2009). Genotyping and downstream filtering was performed with the pipeline JeDi, resulting in a dataset of 706 individuals scored for 8,236,685 total sites, including 51,900 variant sites (including 46,973 biallelic and 4,927 tri- or tetra-allelic).

This dataset includes 1) a tab-separated file with the first column being individual IDs, and the second column their population, 2) mapped reads in a BAM file per individual named “ID.bam”, archived as .tar.gz, which were used as input for the pipeline JeDi.