Identification of eQTL from porcine muscle and liver mRNA sequencing

Author(s)

Felipe André Oliveira Freitas

• Document type: Master's Dissertation
• Press: Piracicaba.
• Institution: Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC)
• Defense date: 2023-05-02

Examining board members

Aline Silva Mello Cesar; Luiz Fernando Brito; Bárbara Silva Vignato

• Advisor: Aline Silva Mello Cesar

Abstract

This study aimed to identify single nucleotide polymorphisms (SNPs) in the transcriptome of skeletal muscle, liver, and brain tissues of Large White pigs, and to investigate their effect on gene regulation in skeletal muscle tissue and porcine liver tissue. Animals were also genotyped using a commercial array of medium density genotyping (GGP-50K), and the data were included in our analyses. Based on RNA-seq data from 72 male pigs, 74,812, 50,932 and 117,330 SNPs were identified in skeletal muscle, liver, and brain tissues, respectively, and also, 38,823 SNPs from the GGP-50K genotyping. The identified variants were annotated, and the effect predicted, resulting in 954 deleterious SNPs that were used for analysis of association with blood biochemical parameters. However, only two deleterious SNPs significantly associated (FDR<0.05) with only the biochemical parameter of blood triglycerides (rs705857876 and rs318362568) were identified. Then, the SNPs from tissue sequencing and GGP-50K genotyping were combined in 4 different scenarios and, applying pruning for Linkage Disequilibrium (LD, r2<0.7), resulting in 4 other scenarios. The 8 scenarios were analyzed to identify local (cis, up to 1Mb) and distal (trans) quantitative expression characteristic (eQTL) loci. There was no significant eQTL (FDR<0.01) for scenarios containing SNPs only from the GGP-50K genotyping, however, when combined with SNPs from skeletal muscle sequencing, significant cis- and trans-eQTLs (FDR<0.01) were identified, associated with expression level of unidentified genes in other scenarios. The number of eQTLs identified in the non-pruning scenarios for LD ranged from 2066 to 2247 in cis- and 43 to 379 in trans. In LD-pruning scenarios (r2>0.7), the eQTL count ranged from 223 to 612 in cis- and from 29 to 403 in trans (FDR<0.01). The total number of genes modulated by eQTLs in scenarios without pruning for LD ranged from 159 to 304 in cis, whereas in trans it ranged from 8 to 1,965. Furthermore, in the LD-pruned scenarios, the total genes ranged from 109 to 185 in cis- and from 6 to 5,993 in trans. The cis- and trans-eQTLs found were not significantly associated (FDR<0.05) with the pig performance traits tested. However, the eQTLs were enriched for several production traits in pigs, thus indicating the overlap of our findings with previous studies. It is noteworthy that the scenario formed only of SNPs from skeletal muscle sequencing and with pruning for LD, presented a greater number of modulated genes, in addition to the greater overlap with the results of the other scenarios. The scenarios constituted by SNPs from the sequencing of both tissues combined with the SNPs from the GGP- 50K genotyping, allowed the identification of cis- and trans-eQTLs that modulate genes (FDR<0.01) that were not identified in the other scenarios. In summary, this study identified a large number of SNPs in the skeletal muscle, liver, and brain tissues of Large White pigs, in addition, the eQTLs found were enriched for swine production traits and in addition to presenting associations with genes that modulate the gene expression of complex traits of pigs. The findings of this study highlight the importance of identifying SNPs in studies of the effects of genotypic variation on the modulation of gene expression and provide valuable information for future studies of genetic selection and pig breeding. (AU)

• FAPESP's process: 21/01694-2 - eQTLl identification from muscle and pig liver RNA sequencing
• Grantee: Felipe André Oliveira Freitas
• Support Opportunities: Scholarships in Brazil - Master