Authors: Da Fonseca, Rute;

Genetic diversity is widely used to understand general processes in evolutionary biology. When applied in biodiversity monitoring and conservation, its absolute value can be converted into risk of extinction. In the context of food production, including agriculture, animal farming and fisheries, low levels of genetic diversity are seen as associated with a higher risk of food insecurity because of e.g. higher susceptibility to disease. Most studies to date have assessed genetic diversity in different species by using a small number of molecular markers (mitochondrial loci and microsatellites) and, more recently, SNP panels. Direct comparisons between such studies are challenging because of ascertainment bias, differences in the sampling design, and variation caused by the specific evolutionary history of the markers, which does not always reflect the overall history of the population/species. Using whole genome sequencing data can potentially provide a platform for unbiased results allowing for comparisons across independent studies. However, the varying quality and availability of genome assemblies across the tree of life requires careful bioinformatics filtering of both assemblies and resequencing data for reproducible results. Here I will discuss the impact of different filtering strategies on the assessment of genetic diversity using whole genome data of non-model organisms.