Contributions to diversity rather than basic measures of genetic diversity characterise the spreading of donkey throughout the American continent

Donkey was introduced into the Americas soon after its discovery in the 15th century. However, there is no historical consensus on how they spread across the continent. In a previous study, two distinct genetic pools (Clusters A -Southern part - and B - Northern part of South America and Central America) were identified, with likely confluence in Colombia. The aim of this study was to evaluate whether the main genetic diversity parameters, such as gene diversity (GD) and allelic richness (k), or the relative contributions of various breeds to these parameters are useful indicators to give genetic support to historical information on putative routes of the spreading of donkeys across the American continent. In full agreement with historical sources suggesting that Greater Antilles were the first breeding nucleus, both total contributions to gene diversity (gGDT) and to allelic richness (CT(k)) showed a higher ability to identify the “abundant centre” of the species on the Continent. Even though there are historical reports suggesting various entry points of the donkey into the continent (e.g. in Brazil), these parameters suggested that, in our dataset, the Cuban donkey population was the more likely representative of the first breeding nucleus of the species. Central and South American donkey populations in the surroundings of the Caribbean Gulf would more likely be early derivatives of Antillean donkey. The strong North-South genetic structure was confirmed for the American donkey metapopulation. Current analyses suggest that populations classified into Cluster A (South) are essentially a sample of the genetic background of Cluster B (North). The Andean route had the highest importance in the formation of the South American populations. The extinction of either population belonging to Cluster B could lead to a decrease in overall genetic diversity both at the gene diversity level (negative gGDT values) and the allelic richness level (positive CT(k) contributions). The opposite pattern is found for populations belonging to Cluster A. The extinction of the populations belonging to Cluster B would decrease the overall American donkey gene diversity in roughly 8% and would dramatically affect the number of alleles in the metapopulation (19.1%). However, the extinction of the donkey populations classified into Cluster A would increase overall gene diversity by 2.2%. Although, the genetic scenario of each individual population varies substantially, the joint conservation of the donkey populations classified into both Clusters A and B is highly advised