In breeding ducks, obtaining the pose information is vital for perceiving their physiological health, ensuring welfare in breeding, and monitoring environmental comfort. This paper proposes a pose estimation method by combining HRNet and CBAM to achieve automatic and accurate detection of duck's multi-poses. Through comparison, HRNet-32 is identified as the optimal option for duck pose estimation. Based on this, multiple CBAM modules are densely embedded into the HRNet-32 network to obtain the pose estimation model based on HRNet-32-CBAM, realizing accurate detection and correlation of eight keypoints across six different behaviors. Furthermore, the model's generalization ability is tested under different illumination conditions, and the model's comprehensive detection abilities are evaluated on Cherry Valley ducklings of 12 and 24 days of age. Moreover, this model is compared with mainstream pose estimation methods to reveal its advantages and disadvantages, and its real-time performance is tested using images of 256 × 256, 512 × 512, and 728 × 728 pixel sizes. The experimental results indicate that for the duck pose estimation dataset, the proposed method achieves an average precision (AP) of 0.943, which has a strong generalization ability and can achieve real-time estimation of the duck's multi-poses under different ages, breeds, and farming modes. This study can provide a technical reference and a basis for the intelligent farming of poultry animals.
Japanese encephalitis virus (JEV) infection alters microRNA (miRNA) expression in the central nervous system (CNS). However, the mechanism contributing to miRNA regulation in the CNS is not known. We discovered global degradation of mature miRNA in mouse brains and neuroblastoma (NA) cells after JEV infection. Integrative analysis of miRNAs and mRNAs suggested that several significantly downregulated miRNAs and their targeted mRNAs were clustered into an inflammation pathway. Transfection with miRNA 466d-3p (miR-466d-3p) decreased interleukin-1β (IL-1β) expression and inhibited JEV replication in NA cells. However, miR-466d-3p expression increased after JEV infection in the presence of cycloheximide, indicating that viral protein expression reduced miR-466d-3p expression. We generated all the JEV coding proteins and demonstrated NS3 helicase protein to be a potent miRNA suppressor. The NS3 proteins of Zika virus, West Nile virus, and dengue virus serotype 1 (DENV-1) and DENV-2 also decreased miR-466d-3p expression. Results from helicase-blocking assays and in vitro unwinding assays demonstrated that NS3 could unwind pre-miR-466d and induce miRNA dysfunction. Computational models and an RNA immunoprecipitation assay revealed arginine-rich domains of NS3 to be crucial for pre-miRNA binding and degradation of host miRNAs. Importantly, site-directed mutagenesis of conserved residues in NS3 revealed that R226G and R202W reduced the binding affinity and degradation of pre-miR-466d. These results expand the function of flavivirus helicases beyond unwinding duplex RNA to degrade pre-miRNAs. Hence, we revealed a new mechanism for NS3 in regulating miRNA pathways and promoting neuroinflammation.IMPORTANCE Host miRNAs have been reported to regulate JEV-induced inflammation in the CNS. We found that JEV infection could reduce expression of host miRNA. The helicase region of the NS3 protein bound specifically to miRNA precursors and could lead to incorrect unwinding of miRNA precursors, thereby reducing the expression of mature miRNAs. This observation led to two major findings. First, our results suggested that JEV NS3 protein induced miR-466d-3p degradation, which promoted IL-1β expression and JEV replication. Second, arginine molecules on NS3 were the main miRNA-binding sites, because we demonstrated that miRNA degradation was abolished if arginines at R226 and R202 were mutated. Our study provides new insights into the molecular mechanism of JEV and reveals several amino acid sites that could be mutated for a JEV vaccine.