Introduction
Feed efficiency is a cornerstone of sustainable aquaculture, with the potential to reduce both production costs (Iversen et al., 2020) and environmental impact (Winther et al., 2020). However, direct phenotyping of individual feed intake and nutrient utilization in Atlantic salmon remains challenging at commercial breeding scales (de Verdal et al., 2017; Knap and Kause, 2018). To phenotype feed efficiency it is necessary to measure individual feed intake and calculate feed efficiency or identify traits in the bioenergetic pathway which contribute to variation in feed efficiency. King among bioenergetic traits are lipid deposition traits, lipid composition traits and their changes over time.
We present a suite of high-throughput, and scalable phenotyping methods aimed at capturing variation in feed intake, lipid metabolism, and overall feed efficiency across life stages. The objective of this study was to evaluate the feasibility, and genetic potential of these methods for phenotyping individual feed intake in Atlantic salmon.
Materials and Methods
We evaluated five novel phenotyping methods across ~2,000 Atlantic salmon from 23 full-sib families. Trials were conducted at the Norwegian Centre for Fish Trials (Ås, Norway) during the freshwater phase (50–100 g) and at Nofima’s Centre for Recirculation in Aquaculture (Sunndalsøra, Norway) during the saltwater phase (500–2,000 g). On three separate occasions, fish were fed experimental diets containing 5% embedded metal micro-pellets and scanned using a calibrated metal detector.
During each trial, fish were fed to satiation in small family groups and scanned post-prandially using a semi-portable conveyor-based metal detector. At the third and final sampling, fish were euthanized humanely, gutted, and filleted. Feed intake was then confirmed through gut analysis using both metal detection and X-ray imaging. In parallel, tissue samples were collected for stable isotope and spectral analyses.
The data was analyzed with a linear mixed animal model using the DMU software (Madsen and Jensen, 2013), including fixed tank and sex, and a random additive genetic effect based on the genomic relationship matrix (G). Heritability (h²) was estimated from variance components. The five phenotyping approaches evaluated were:
Stable Isotope Analysis: δ¹³C and δ¹⁵N values were analyzed in fillet tissue to infer carbon and nitrogen assimilation as indicators of nutrient utilization and feed efficiency (Dvergedal et al., 2019, 2020). Family-level isotope ratios showed correlation with feed conversion ratio, providing a biochemical proxy trait for breeding. Metal Detector Scanning: A calibrated, conveyor-driven metal detector was used to non-invasively quantify individual feed intake based on postprandial field strength signals. The metal beadlets are Fish radio-opaque and were X-rayed post gutting to count markers in the gastrointestinal tract. Individual feed intake was estimated by calibrating marker count to feed weight (Difford et al., 2023; Ahmad et al., 2025b). Interactance Near-Infrared Spectroscopy (iNIR): Whole-body, fillet, and viscera NIR spectra were used to predict lipid content and feed intake through partial least-squares regression (PLSR) (Ahmad et al., 2025a). Raman Spectroscopy: Portable Raman probes were applied to fillets to assess fatty acid composition, focusing on omega-3 content (Lintvedt et al., 2022). This method is non-destructive and shows promise for high-throughput lipid phenotyping in breeding programs.
Results
Preliminary results from these trials will be presented. The study demonstrates the feasibility of several novel, non-invasive methods for phenotyping feed intake and efficiency in Atlantic salmon. While each method shows potential for application in selective breeding, differences in accuracy, heritability, and scalability were observed. These findings highlight both the promise and the current limitations of emerging phenotyping technologies, underscoring the need for continued research and development to improve their reliability, biological interpretation, and technology readiness level.
Funding
This study was supported by the Norwegian research council project DigiFishent (334821) and Foods of Norway, WP5 (237841).
References
Ahmad, A., J. Petter, A. Kristina, B. Hatlen, B. Sime, P. Berg, A. Norris, and G. Frank. 2025a. Genetic and phenotypic validation of whole body fat content measured across production phases of Atlantic salmon using dielectric and near infrared Interactance spectroscopy. Aquaculture 596:741747. doi:10.1016/j.aquaculture.2024.741747.
Ahmad, A., A.K. Sonesson, B. Hatlen, G. Bæverfjord, P. Berg, A. Norris, and G.F. Difford. 2025b. Genetic analysis of individual feed intake and efficiency in Atlantic salmon smolts using X-ray imaging. Aquaculture 608:742715. doi:10.1016/j.aquaculture.2025.742715.
Difford, G.F., B. Hatlen, K. Heia, G. Bæverfjord, B. Eckel, K.H. Gannestad, O.H. Romarheim, S. Lindberg, A.T. Norris, A.K. Sonesson, B. Gjerde, and L. Weber. 2023. Digital phenotyping of individual feed intake in Atlantic salmon ( Salmo salar ) with the X-ray method and image analysis 1–11. doi:10.3389/fanim.2023.1177396.
Dvergedal, H., J. Ødegård, L.T. Mydland, M. Øverland, J.Ø. Hansen, R.M. Ånestad, and G. Klemetsdal. 2019. Stable isotope profiling for large-scale evaluation of feed efficiency in Atlantic salmon (Salmo salar). Aquac. Res. 50:1153–1161. doi:10.1111/are.13990.
Dvergedal, H., L. Torunn Mydland, and G. Klemetsdal. 2020. The change in 15N stable isotope content in muscle, liver and mid-intestine in juvenile Atlantic salmon (Salmo salar) under starvation. Aquac. Res. 51:5265–5268. doi:10.1111/are.14840.
Iversen, A., F. Asche, Ø. Hermansen, and R. Nystøyl. 2020. Production cost and competitiveness in major salmon farming countries 2003–2018. Aquaculture 522:735089. doi:10.1016/j.aquaculture.2020.735089.
Knap, P.W., and A. Kause. 2018. Phenotyping for genetic improvement of feed efficiency in fish: Lessons from pig breeding. Front. Genet. 9:1–10. doi:10.3389/fgene.2018.00184.
Lintvedt, T.A., P.V. Andersen, N.K. Afseth, B. Marquardt, L. Gidskehaug, and J.P. Wold. 2022. Feasibility of in-line Raman spectroscopy for food quality assessment in food industry – how fast can we go?. Appl. Spectrosc.. doi:10.1177/00037028211056931.
Madsen, P., and J. Jensen. 2013. A user’s guide to DMU. Cent. Quant. Genet. Genomics Dept. Mol. Biol. Genet. Univ. Aarhus Res. Cent. Foulum Box 50, 8830 Tjele Denmark 1–32.
de Verdal, H., H. Komen, E. Quillet, B. Chatain, F. Allal, J.A.H. Benzie, and M. Vandeputte. 2017. Improving feed efficiency in fish using selective breeding: a review. Rev. Aquac. 10:1–19. doi:10.1111/raq.12202.
Winther, U., E.S. Hognes, S. Jafarzadeh, and F. Ziegler. 2020. Greenhouse Gas Emissions of Norwegian Seafood Products in 2017.