Genetic analyses of novel traits derived from CT scanning for implementation in terminal sire sheep breeding programmes

This study used archived CT scanning images from Texel (n= 3534), Suffolk (n= 2357) and Charollais (n= 2013) ram lambs, scanned between 1997 and 2016 as part of the national UK terminal sire breeding programmes, to measure new, previously unexploited, CT phenotypes for product quality. The new CT phenotypes included 12 traits relating to eye muscle dimensions, spine region lengths and vertebrae numbers, and intramuscular fat (IMF), as a predictor of meat quality. The aim was to estimate genetic parameters for these new traits, confirm genetic parameters for existing growth and carcass traits, and investigate relationships amongst traits, when adjusted for either age or live weight. The new CT traits were generally moderately heritable (h2 ranging from 0.23-0.54 when adjusted for age) within each breed. Notable exceptions were: vertebrae counts, which had lower heritabilities (0.11-0.18) in the Suffolk and Texel; lumbar region spine length and vertebrae count in the Charollais, (0.04); and CT-predicted IMF in the Charollais (0.88). Generally, genetic variation decreased, but heritabilities increased, when traits were adjusted for live-weight rather than age, ranging from h2 0.21-0.74 for the new CT traits, except for the Charollias lumbar spine length (0.01) and CT-predicted IMF (0.87). Within-breed genetic correlations between new CT traits and current CT traits (carcass fat and muscle weights, muscularity), ultrasound tissue depths, or scan weight (at ~21 weeks), were low to moderate. High genetic correlations were observed between ultrasound and CT measured eye muscle depth (r = 0.77-0.85). Genetic correlations between CT-predicted IMF and total carcass fat estimates, however, were unfavourable (large and positive; r = 0.66-0.96), especially in the Suffolk and Charollais, suggesting little scope for divergent selection between fat depots. Some unfavourable genetic correlations of spine length with CT muscle weight or eye muscle depth were also estimated, after live weight adjustment. Possibilities therefore exist for genetic selection on additional product quality traits that can be routinely measured by CT scanning. Adjusting new and existing carcass traits for live weight, rather than age, should increase heritabilities and provide a more relevant industry end-point. These findings could contribute to the production of a new selection index, with optimal weightings on growth, carcass and product quality traits, for a sustainable UK terminal sire sheep breeding programme