Variabilidad nucleotídica y estructura poblacional en cerdos criollos cubanos

The Cuban Creole pig constitutes a valuable genetic resource that has been integrated by more than 500 years to history, culture and the living of the Cuban farmer. Its high rusticity turns into an important food supply and income without additional investment. The objectives of the present work are the description of the levels and patterns of nucleotide variation using the Control (D-loop) and Cytochrome b (CytB) regions of the mitochondrial DNA (mitDNA), as well as the use of coalescent simulations to establish reasonable demographic models that explain the observed data. A total of 32 samples from 6 localities across the country were sequenced. The nucleotide and haplotype diversities in the island are in the order of the values observed in Europe, although inferiors to Asia. Population structure was detected between three different geographic regions of the island (West, Center and East). Median-joining network analysis detected great haplotype similarities with the European populations and in one case a haplotype near the Asian populations. However, the number of exclusive haplotypes in the Cuban population was very high. Apparently, the observed variability has different oirgins according to the region. In the West, it seems to be the result of migratory processes, whereas in the East these processes seem less important. We proposed 4 demographic models based on the most significant historical events of the island. Each model introduces an additional demographic event to clarify the importance of these events in the present variability of the populations. By coalescence simulations and Approximate Bayesian Computation approach based on descriptive statistics, we obtained the compound probabilities of each of the models, which were compared by means of Akaike information criterion methodology. The simplest model could explain the results, being the effective population size greater on the West, and the smaller on the Center. Although this model contradicts in certain ways some of the results, this could be due to the lack of power of this analysis to discriminate between the models