Unravelling novel and closely linked association signals for fat-related traits in pigs using prioritised variants...

For most production traits, the largest proportions of genetic variance remain unmapped. Dense whole-genome sequence (WGS) data enable the possibility of discovering novel associations as well as unravelling closely linked association signals with a resolution that marker arrays cannot reach. However, the identification of variants from WGS data that are causal of the variation of complex traits is hindered by the high dimensionality and linkage disequilibrium. Thus, at best, we can narrow the circle around the causal variants to prioritise a set of variants for their posterior validation. In this study, we assessed the utility of WGS data for uncovering associations of weaker effects using, as a model, fat content and composition traits in a Duroc pig population where we previously described major effects of the LEPR and SCD genes. We genotyped 971 pigs for a set of 182 variants from 154 candidate genes that were prioritised from amongst the WGS variants discovered in 205 sequenced individuals. These variants were prioritised conditional to LEPR and SCD. The association of the prioritised variants with the target traits was then tested in the confirmation set of 971 pigs. A total of 17 potentially independent quantitative trait loci (8.4% of the total number of studied genes) were significantly associated (q-value < 0.05) with at least one of the studied traits. We identified novel associations attributable to genes such as ABCC2, MOGAT2, or PLPP1 for backfat thickness, myristic acid content, and monounsaturated fatty acid content, respectively. Our results also revealed a finer granularity of weaker genetic effects in loci such as those around the DGAT2 and FADS2 genes, which may mask the effects of closely located genes like MOGAT2 and DAGLA, respectively. To refine the prioritisation of variants for validation studies, especially when targeting those of weaker effects, we recommend larger and more diverse discovery sets, more precise and complete functional gene annotation, and the integration of other omics data.