Identification and Genome Characterization of a Novel Nege-like Virus Isolated from Aphids (Aphis gossypii) in Yunnan Province.

Negeviruses are insect-specific enveloped RNA viruses that exhibit a wide geographic distribution. A novel nege-like virus, tentatively named Aphis gossypii nege-like virus (AGNLV, GenBank: OR880429.1), was isolated from aphids (Aphis gossypii) in Lijiang City, Yunnan, China. AGNLV has a genome sequence of 9258 nt (excluding the polyA tail) encoding three open reading frames (ORFs). ORF1 (7149 nt) encodes a viral methyltransferase, a viral RNA helicase, and an RNA-dependent RNA polymerase. ORF2 (1422 nt) encodes a DiSB-ORF2_chro domain and ORF3 encodes an SP24 domain. The genome sequence of AGNLV shares the highest nucleotide identity of 60.0% and 59.5% with Wuhan house centipede virus 1 (WHCV1) and Astegopteryx formosana nege-like virus (AFNLV), respectively. Phylogenetic analysis based on the RNA-dependent RNA polymerase shows that AGNLV is clustered with other negeviruses and nege-like viruses discovered in aphids, forming a distinct "unclassified clade". Interestingly, AGNLV only encodes three ORFs, whereas AFNLV and WHCV1 have four ORFs. Structure and transmembrane domain predictions show the presence of eight alpha helices and five transmembrane helices in the AGNLV ORF3. Translational enhancement of the AGNLV 5' UTR was similar to that of the 5' UTR of plant viruses. Our findings provide evidence of the diversity and structure of nege-like viruses and are the first record of such a virus from a member of the genus Aphis.

Open-Set Recognition of Individual Cows Based on Spatial Feature Transformation and Metric Learning.

The automated recognition of individual cows is foundational for implementing intelligent farming. Traditional methods of individual cow recognition from an overhead perspective primarily rely on singular back features and perform poorly for cows with diverse orientation distributions and partial body visibility in the frame. This study proposes an open-set method for individual cow recognition based on spatial feature transformation and metric learning to address these issues. Initially, a spatial transformation deep feature extraction module, ResSTN, which incorporates preprocessing techniques, was designed to effectively address the low recognition rate caused by the diverse orientation distribution of individual cows. Subsequently, by constructing an open-set recognition framework that integrates three attention mechanisms, four loss functions, and four distance metric methods and exploring the impact of each component on recognition performance, this study achieves refined and optimized model configurations. Lastly, introducing moderate cropping and random occlusion strategies during the data-loading phase enhances the model's ability to recognize partially visible individuals. The method proposed in this study achieves a recognition accuracy of 94.58% in open-set scenarios for individual cows in overhead images, with an average accuracy improvement of 2.98 percentage points for cows with diverse orientation distributions, and also demonstrates an improved recognition performance for partially visible and randomly occluded individual cows. This validates the effectiveness of the proposed method in open-set recognition, showing significant potential for application in precision cattle farming management.