Accurate body dimension and weight measurements are critical for optimizing poultry management, health assessment, and economic efficiency. This study introduces an innovative deep learning-based model leveraging multimodal data—2D RGB images from different views, depth images, and 3D point clouds—for the non-invasive estimation of duck body dimensions and weight. A dataset of 1023 Linwu ducks, comprising over 5000 samples with diverse postures and conditions, was collected to support model training. The proposed method innovatively employs PointNet++ to extract key feature points from point clouds, extracts and computes corresponding 3D geometric features, and fuses them with multi-view convolutional 2D features. A Transformer encoder is then utilized to capture long-range dependencies and refine feature interactions, thereby enhancing prediction robustness. The model achieved a mean absolute percentage error (MAPE) of 5.73% and an R² of 0.953 across seven morphometric parameters describing body dimensions, and an MAPE of 10.49% with an R² of 0.952 for body weight, indicating robust and consistent predictive performance across both structural and mass-related phenotypes. Unlike conventional manual measurements, the proposed model enables high-precision estimation while eliminating the necessity for physical handling, thereby reducing animal stress and broadening its application scope. This study marks the first application of deep learning techniques to poultry body dimension and weight estimation, providing a valuable reference for the intelligent and precise management of the livestock industry with far-reaching practical significance.

Skeletal muscle satellite cells (SMSCs) are essential for embryonic myogenesis and postnatal muscle regeneration; however, their cellular heterogeneity and transcriptional dynamics during avian development remain largely unexplored. Here, we performed single-cell RNA sequencing (scRNA-seq) on 42,886 cells isolated from goose leg muscles across four embryonic stages (E13, E15, E18, and E23), with each stage comprising pooled tissues from four female embryos. Unbiased clustering resolved 22 transcriptionally distinct clusters representing six major cell types—satellite cells, myocytes, fibro-adipogenic progenitors, endothelial cells, immune cells, and Schwann cells—with satellite cells being the most abundant. Satellite cells were further subdivided into three functional states (quiescent, activated, and proliferative/differentiating), which followed a continuous, linear pseudotime trajectory from early to late embryonic stages. This trajectory was marked by a progressive downregulation of stemness-associated regulators (e.g., PAX7) and upregulation of myogenic commitment and differentiation factors (e.g., MYF5, MYOD1, and MYOG), faithfully mirroring chronological development. Cell–cell communication analysis revealed that quiescent satellite cells exhibited the most extensive intercellular signaling networks (e.g., FGFR, Ephrin, collagen, CADM), whereas activated and proliferative/differentiating cells showed progressively diminished communication capacity. Across developmental stages, the contribution intensities of key signaling pathways—including SEMA6, CDH, FGF, LAMININ, MK, MPZ, CADM, FN1, and COLLAGEN—varied significantly among satellite cell states, indicating state-specific responsiveness to microenvironmental cues. Collectively, these findings demonstrate that satellite cells dynamically coordinate extrinsic signal integration with intrinsic differentiation programs to achieve orderly myogenic progression. This study provides a high-resolution single-cell atlas of goose SMSC development, uncovering subpopulation heterogeneity, state-specific molecular signatures, and key signaling pathways, with important implications for avian muscle biology and genetic improvement of poultry.

Wugang Copper Goose is an indigenous meat-type breed, but the molecular events accompanying leg muscle development around hatch remain unclear. We examined breast and leg muscle histology at embryonic days 15 (E15), 19 (E19), 23 (E23), and 28 (E28) and posthatch day 1 (P1). Transcriptomic, proteomic, and metabolomic profiling of leg muscle at E15, E23, and P1 used matched samples from the same five individuals per stage. Leg muscle fibers became progressively organized before hatch and enlarged markedly at P1, and multi-omics analysis distinguished two developmental transitions. From E15 to E23, increasing cytoskeletal and calcium-signaling programs accompanied lower DNA-replication- and nucleotide-related signals, consistent with structural assembly and declining replication-related activity. From E23 to P1, fiber enlargement coincided with higher oxidative phosphorylation, citrate-cycle, peroxisome proliferator-activated receptor signaling, and fatty-acid-associated programs and lower proliferative programs. RNA and protein fold changes were moderately concordant across both transitions (Spearman ρ = 0.55 and 0.58), whereas metabolite changes were largely layer specific. Reverse-transcription quantitative PCR (RT-qPCR) supported the directional expression patterns of eight representative genes. Together, these data distinguish an early transition from replication-related activity to structural assembly and a posthatch transition dominated by oxidative and lipid-associated maturation.

Yan goose, a representative Chinese local breed, requires genetic resource conservation. To address this, we optimized a semen cryopreservation protocol for Yan geese and validated its feasibility in large-scale cryobank production. Using motility of frozen-thawed sperm (MFS) as the primary indicator, supplemented by artificial insemination (AI) trials, we systematically evaluated cryoprotectant type, freezing rate, extender osmolality and pH, and thawing rate. Dimethylformamide (DMF) significantly outperformed dimethylacetamide (DMA) in MFS (34.17% vs. 24.17%, p < 0.05). Based on optimization data and practical requirements, the final DMF method comprised freezing at 2.0 cm above liquid nitrogen for 2 min, extender osmolality of 335 mOsm/kg at pH 7.8, and thawing at 5 °C for 60 s. Teaser-induced semen collection was successfully applied in a Chinese local goose breed for the first time, achieving a 69.14% success rate. Using this system, production over 17 days yielded 1225 straws (96.61% qualified), of which 838 from 30 families were banked. AI of banked semen achieved a fertilization rate of 70.4% and a hatchability of 86.8%. These findings demonstrate the successful establishment of a cost-effective cryobank for Yan geese, providing a feasible paradigm for conserving other Chinese local goose breeds.