Genome-wide association study (GWAS) for residual feed intake (RFI) and component traits of feed efficiency in pigs divergently selected for RFI

Abstract Text:

The objective of this work was to identify genomic regions associated with residual feed intake (RFI, defined as observed minus expected feed intake) and component traits of feed efficiency in 2 lines of pigs selected for high (H-RFI) and low (L-RFI) RFI. Phenotypic and genotypic data on 994 L-RFI pigs from generations 0 to 10 and 698 H-RFI pigs from generations 4 to 10 were used. Phenotypic data included average daily feed intake (ADFI) and ADG from ~40 to 118 kg, backfat depth (BF) and loin muscle area (LMA) at off-test, feed conversion ratio (feed-to-gain; FCR), and RFI (ADFI adjusted for ADG, BF, and metabolic BW). All animals were genotyped using the Illumina PorcineSNP60 BeadChip. After quality control, 51,098 SNPs were used for analyses. Genome-wide associations were performed one SNP at a time as a fixed-effect (combined test for additive and dominance effects) in an animal model. Models also included the fixed effects of sex and contemporary group. Appropriate covariates were fitted for each trait analyzed. Analyses were performed separately for each line and jointly. SNP associations were deemed significant based on Bonferroni-adjustment when –log₁₀(P)>6. The total number of significant associations using data on L-RFI, H-RFI, and the joint lines (L-RFI/H-RFI/joint) for each trait were, ADFI: 0/0/32, ADG: 0/87/119, BF: 12/0/15, FCR: 5/2/4, LMA: 0/4/0, and RFI: 1/1/0. All associations with ADFI were located on SSC1, between 168-179 Mb. All associations for ADG were found near this same region on SSC1 (154-180 Mb), with the exception of 3 unmapped. One SNP in this region was also associated with FCR, in addition to associations found on SSC2, 6, 7 and 8. BF showed the most significant association in this study (joint: -log₁₀(P)=23.1), with a QTL on SSC2 (162 Mb). Other associated regions for BF were found on SSC1, 2, 7, 8, 9 and 17. All 4 associations for LMA were found in the H-RFI line (on SSC12 and 16). Each RFI line had a significant association for RFI, on SSC2 (67 Mb; L-RFI) and SSC6 (30 Mb; H-RFI). Results indicate that regions associated with RFI and component traits of feed efficiency might (ADFI and ADG) or might not (BF, FCR, LMA, and RFI) be the same across lines of pigs divergently selected for RFI. Financial support from AFRI-NIFA (No. 2011-68004-30336) is appreciated.

Keywords: genomics, QTL mapping, swine