Bayesian GWAS and network analysis revealed new candidate genes for number of teats in pigs

• Autor(a) principal: Verardo, L. L.
• Data de Publicação: 2014
• Outros Autores: Silva, F. F., Varona, L., Resende, M. D. V., Bastiaansen, J. W. M., Lopes, P. S., Guimarães, S. E. F.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://dx.doi.org/10.1007/s13353-014-0240-y; http://www.locus.ufv.br/handle/123456789/19820
• topic: Counting data; Genes; Reproductive traits; SNP association

The genetic improvement of reproductive traits such as the number of teats is essential to the success of the pig industry. As opposite to most SNP association studies that consider continuous phenotypes under Gaussian assumptions, this trait is characterized as a discrete variable, which could potentially follow other distributions, such as the Poisson. Therefore, in order to access the complexity of a counting random regression considering all SNPs simultaneously as covariate under a GWAS modeling, the Bayesian inference tools become necessary. Currently, another point that deserves to be highlighted in GWAS is the genetic dissection of complex phenotypes through candidate genes network derived from significant SNPs. We present a full Bayesian treatment of SNP association analysis for number of teats assuming alternatively Gaussian and Poisson distributions for this trait. Under this framework, significant SNP effects were identified by hypothesis tests using 95 % highest posterior density intervals. These SNPs were used to construct associated candidate genes network aiming to explain the genetic mechanism behind this reproductive trait. The Bayesian model comparisons based on deviance posterior distribution indicated the superiority of Gaussian model. In general, our results suggest the presence of 19 significant SNPs, which mapped 13 genes. Besides, we predicted gene interactions through networks that are consistent with the mammals known breast biology (e.g., development of prolactin receptor signaling, and cell proliferation), captured known regulation binding sites, and provided candidate genes for that trait (e.g., TINAGL1 and ICK).