Genetic parameters and genome-wide association studies of image-derived cardiac traits in Tasmanian Atlantic salmon (Salmo salar)

Abstract

Cardiac morphology, including ventricle shape and epicardial fat coverage, differs significantly between farmed and wild Atlantic salmon (Salmo salar) and is associated with health and performance. While environmental and dietary effects are increasingly understood, the genetic basis of these traits remains less explored in part due to the difficulty in obtaining phenotypes. We developed an image-based phenotype for epicardial fat coverage, achieving a strong correlation with ground truth measurements obtained using computer tomography (R = 0.87). An algorithm to measure ventricle height and width from images was also developed. Using data collected from 1792 individuals from the Salmon Enterprises of Tasmania (SALTAS) selective breeding program, we performed quantitative genetic and genomic analyses of these traits, along with heart weight and cardio somatic index (CSI). Estimated heritabilities were 0.46 ± 0.06 for epicardial fat coverage, 0.33 ± 0.05 for heart weight, 0.30 ± 0.05 for CSI, and 0.29 ± 0.05 for ventricle shape. Genetic correlations between cardiac traits were weak to moderate (0.04–0.51), as were all correlations between cardiac and production traits (0.00–0.46) except for that between heart weight and harvest weight (0.85 ± 0.04). Genome wide association studies demonstrated that all cardiac traits are highly polygenic and identified candidate genes linked to cardiac function, cardiovascular development, lipid metabolism, and obesity. These results highlight the potential of image-based phenotyping for selective breeding and suggest genetic improvement or monitoring of cardiac morphology and health is achievable. However, further studies are needed to validate these findings and better understand associations between cardiac traits, health, and performance.