Pig carcass traits are among the most economically significant characteristics and are crucial for genetic selection in breeding and enhancing the economic efficiency. Standardized and automated carcass phenotyping can greatly enhance the measurement efficiency and accuracy, thereby facilitating the selection and breeding of superior pig carcasses. In this study, we utilized phenotypic images and data from 3912 pigs to propose a deep learning-based approach for the automated determination of pig carcass phenotypic traits. Using the YOLOv8 algorithm, our carcass length determination model achieves an average accuracy of 99% on the test set. Additionally, our backfat segmentation model, YOLOV8n-seg, demonstrates robust segmentation performance, with a Mean IoU of 89.10. An analysis of the data distribution comparing manual and model-derived measurements revealed that differences in the carcass straight length are primarily concentrated between −2 cm and 4 cm, while differences in the carcass diagonal length are concentrated between −3 cm and 2 cm. To validate the method, we compared model measurements with manually obtained data, achieving coefficients of determination (R2) of 0.9164 for the carcass straight length, 0.9325 for the carcass diagonal length, and 0.7137 for the backfat thickness, indicating high reliability. Our findings provide valuable insights into automating carcass phenotype determination and grading in pig production.
Background: Lean meat percentage (LMP), longissimus muscle area (LMA), and longissimus muscle depth (LMD) are critical carcass traits that directly impact meat yield and the economic efficiency of swine production. Although genome-wide association studies (GWAS) have identified numerous loci associated with these traits, the functional characterization of candidate quantitative trait loci (QTLs) and their regulatory mechanisms is still limited and poses significant challenges. Here, we conducted GWAS in 4593 pigs from three commercial populations (Yorkshire, Landrace and Duroc) to identify loci associated with LMP, LMA and LMD. We further integrated integrative chromatin accessibility, histone modifications, chromatin interaction information, and dual-luciferase reporter assays to identify candidate QTLs with potential regulatory functions. Candidate genes were subsequently prioritized using single-nucleus RNA sequencing, Phenome-wide association analysis, and Mendelian randomization. Results: This multi-layered approach prioritized candidate regulatory QTLs associated with LMP, LMA and LMD, and identified high-confidence candidate genes potentially involved in the regulation of meat yield traits. For LMP, WWOX and ABL1 showed the strongest evidence in Yorkshire and Landrace population, respectively. For LMA, PHB1 and BAZ2A showed the strongest evidence in the Yorkshire and combined-population, respectively. SLC16A7 was specifically associated with both LMP and LMA in Landrace. For LMD, ENSSSCG00000045482 and RSPO4 showed the strongest evidence in Yorkshire pigs, while XPO4 was identified in the combined-population analysis. These genes may influence muscle metabolic homeostasis, muscle proliferation, hypertrophy, and atrophy through interactions with myogenic transcription factors and key signaling pathways such as PPAR-RXR and Wnt/β-catenin. Conclusions: These findings provide new insights into the genetic and regulatory basis of meat yield traits in pigs and highlight candidate regulatory QTLs and genes for future functional validation and precision breeding applications aimed at improving pork yield.