PigPhenomics: metabolomics merged with genomics to develop novel breeding and selection approaches in pigs

Filling the gap between phenotypes and genotypes is crucial for a proper understanding of biological complexity of an organism. This generates more interest when disentangling biological complexity can be directly applied to the development of genetic tools that can guide animal breeding, selection and management decisions for a more sustainable animal-production sector. A comprehensive and systematic collection of phenotypes, as well as the production of genomic data, will help to bridge the gap between genetic variability and the production traits of the animals. Omics techniques have changed the way to investigate the complexity of biological systems. In the last years, the availability of more and more sophisticated high-throughput sequencing genomics, transcriptomics) and phenotyping technologies (proteomics and metabolomics) coupled with the massive amount of produced data, has spawned ‘phenomics’, the use of large-scale approaches to study how genetic instructions translate into the full set of phenotypic traits of an organism. In addressing this challenge, we applied a phenomic approach in two heavy pig breeds. We collected and generated phenotypic data including production traits (endpoint information) and blood metabolites (intermediate phenotypes). Metabolomics information was obtained using both mass spectrometry (MS/MS) and nuclear magnetic resonance (NMR) analytical pipelines. About 1000 small compounds belonging to different biological classes including amino acids, amines, carbohydrates, peptides, lipids and sugars were identified and quantified. Genomic data were obtained from the Illumina Porcine 60K e 70K BeadChips and from whole-genome sequencing data. Phenotypes and genotypes were linked together via univariate and multivariate statistics. Genome scans evidenced many genomic regions controlling metabolite levels as well as effects on production traits. Re-sequencing data were used to identify causative mutations. The metabolic network reconstruction and the related cluster analyses led to identify common and specific features characterizing the different pig breeds/lines. Overall, the obtained results not only can be useful to better understand the genetic architecture of physiological traits in pigs but also represent a first effort to establish genetic tools for precise pig breeding and selection programs.