Genetic estimates of relatedness: Established practices and new opportunities through low coverage whole genome sequencing

Annika Freudiger, Natalie Kestel, Vladimir M. Jovanovic, Mariana Madruga de Brito, Angelina V. Ruiz-Lambides, Katja Nowick, Anja Widdig, Harald Ringbauer

Identifying close relatives in wild animal populations is fundamental across many research fields. Genetic estimates of relatedness have expanded rapidly in recent decades, based on various types of genetic data. Here, we review their use and outline opportunities for future studies by combining two complementary approaches. First, we systematically reviewed 2,861 articles to assess how genetic relatedness has been estimated over time. Second, we compare widely used genetic data types for inferring relatedness, conducting computational experiments using data from a rhesus macaque (Macaca mulatta) population in Puerto Rico. We compared other methods against precise identity-by-descent segment-based estimates of relatedness. Our results show that most studies of relatedness (89%) continue to rely on short tandem repeat (STR) markers, despite their limited precision. Single-nucleotide polymorphism (SNP)- marker-based relatedness estimates remain underused (8.3% of studies), even though they yield more reliable estimates when sampled in sufficient numbers. Finally, we find that the simple pairwise-mismatch rate (PMR) method for estimating relatedness in low-coverage WGS data (commonly used in human ancient DNA studies) works robustly for low-coverage data, e.g., DNA retrieved from faecal samples or from cost-effective low-coverage whole-genome sequencing (lcWGS). Together, our findings highlight lcWGS combined with PMR-based relatedness estimation as a promising, cost-effective alternative when DNA quality is limited, genomic resources are scarce, or economic efficiency is essential.