Geographic and taxonomic biases in the vertebrate tree of life

Aim Resolving the tree of life is among the greatest challenges for modern biology, yet genetic data for many species are lacking to infer their position in the tree with confidence. In fact, little is known about what the missing species are and where to look for them. I identify main hotspots that host the most unsampled vertebrate species and test the hypothesis that local diversity determines the degree of sampling effort. Location Global. Taxon Terrestrial vertebrates (amphibians, birds, mammals, reptiles). Methods I use published distributional data and search DNA sequence repositories to identify taxa with no genetic data available. I map spatial patterns of the unsampled diversity and sampling effort to test the effect of species richness on sampling completeness. I examine the dependence of country-level sampling completeness on the country's socio-economic development. Results About 24% of living terrestrial vertebrate species have no genetic data available. The Andes, Central Africa and the Malay Archipelago are the most unsampled. Sampling effort is highest in North America, continental South-East Asia and Amazonia. In contrast, Central Africa, the Horn of Africa, Malay Archipelago and Andes are consistently the least sampled regions. Regions of low diversity are better sampled than biodiversity hotspots. Main conclusions The most unsampled species are in the tropics. I pinpoint several key regions that are least represented in the tree of life and where sampling should be prioritized. Country's wealth seems to be an informative proxy for its sampling completeness.

Automated summary

This study identifies hotspots with the most unsampled vertebrate species, mainly in tropical regions. It also suggests that the socio-economic development of a country may be indicative of its sampling completeness.