Effect of domestication in the pig genome.

Abstract

Animal domestication is an important process in the human history in which different traits of the animals were selected, such as faster growth or greater docility. To study domestication at the genetic level it is necessary to identify the markers related to this evolutionary process. Advances in sequencing technologies have improved the investigation of the genomics of domestication, which has allowed to determine the genetic changes that cause this transformation from wild to domestic species. The main goal of this thesis is the evaluation of the domestication effect in the pig genome through the analysis of genetic diversity in domestic and wild populations. In the first part, analyses of differentiation and linkage disequilibrium were performed to detect differences between domestic and wild pigs, using the pathway as the unit of analysis. Through the study of differentiation, using the Fst statistic, we obtained significant pathways related to behavior and development, which were some of the first selected traits in pigs. On the other hand, when performing the disequilibrium analysis, using the nSL statistic, we detected differences in pathways related to the reproduction of the animal, a recently selected trait. Besides, we made a co-association network using all pathways that are significantly different between domestic and wild pigs, obtaining three differentiated clusters, one related to growth and hormonal regulation, another with the sympathetic nervous system and the last with the reproduction. In the second part, we performed an analysis of the strength of selection at the genome level in domestic and wild pigs, using two very different domestic populations, Iberian and Large White. Iberian breed is an autochthonous breed that has recently suffered a strong reduction in the effective population size, Large White is an international commercial breed that has been artificially improved and introgressed with Asian pigs. To analyze the strength of the selection we use the parameter α, which estimates the proportion of non-synonymous substitutions that are adaptive, using four different estimators of variability, each focused on a part of the frequency spectrum: Fu&Li (only singletons), Watterson (whole spectrum giving more weight at low frequencies), Tajima (whole spectrum weighted uniformly) and Fay&Wu (increases the weight proportionally with the frequency). However, when analyzing the selection patterns, we did not find more common signals between the two domestic breeds than between domestic and wild ones. Instead, we found a larger effect of demography on the selection, Iberian has a very low variability due to its low population size, which is shown in the obtained selection patterns, which resemble a population reduction; while Large White has a larger variability, possibly due to the presence of Asian alleles in its genome, obtaining patterns that can be explained by the presence of both deleterious and beneficial mutations, together with a population expansion and/or migration. Finally, we have developed a web-based application to analyze VCF files, which can help identify possible errors or biases, mainly related to the SNP coverage.

Date of Award	5 Jul 2019
Original language	English
Supervisor	Sebastián Ernesto Ramos Onsins (Director), Mario Caceres Aguilar (Tutor) & Miguel Perez Enciso (Director)