How animals, particularly livestock, adapt to various climates and environments over short evolutionary time is of fundamental biological interest. Further, understanding the genetic mechanisms of adaptation in indigenous livestock populations is important for designing appropriate breeding programs to cope with the impacts of changing climate. Here we conducted a comprehensive genomic analysis of diversity, interspecies introgression and climate-mediated selective signatures in a global sample of sheep and their wild relatives. By examining 600k and 50k genome-wide SNP data from 3447 samples representing 111 domestic sheep populations and 403 samples from all their seven wild relatives (argali, Asiatic mouflon, European mouflon, urial, snow sheep, bighorn and thinhorn sheep), coupled with 88 whole-genome sequences, we detected clear signals of common introgression from wild relatives into sympatric domestic populations, thereby increasing their genomic diversities. The introgressions provided beneficial genetic variants in native populations, which were significantly associated with local climatic adaptation. We observed common introgression signals of alleles in olfactory-related genes (e.g., ADCY3 and TRPV1) and the PADI gene family including in particular PADI2, which is associated with antibacterial innate immunity. Further analyses of whole-genome sequences showed that the introgressed alleles in a specific region of PADI2 (chr2: 248302667-248306614) correlate with resistance to pneumonia. We conclude that wild introgression enhanced climatic adaptation and resistance to pneumonia in sheep. This has enabled them to adapt to varying climatic and environmental conditions after domestication.

No abstract

No abstract

No abstract

Deep learning-based dense object detectors have achieved great success in the past few years and have been applied to numerous multimedia applications such as video understanding. However, the current training pipeline for dense detectors is compromised to lots of conjunctions that may not hold. In this paper, we investigate three such important conjunctions: 1) only samples assigned as positive in classification head are used to train the regression head; 2) classification and regression share the same input feature and computational fields defined by the parallel head architecture; and 3) samples distributed in different feature pyramid layers are treated equally when computing the loss. We first carry out a series of pilot experiments to show disentangling such conjunctions can lead to persistent performance improvement. Then, based on these findings, we propose Disentangled Dense Object Detector (DDOD), in which simple and effective disentanglement mechanisms are designed and integrated into the current state-of-the-art dense object detectors. Extensive experiments on MS COCO benchmark show that our approach can lead to 2.0 mAP, 2.4 mAP and 2.2 mAP absolute improvements on RetinaNet, FCOS, and ATSS baselines with negligible extra overhead. Notably, our best model reaches 55.0 mAP on the COCO test-dev set and 93.5 AP on the hard subset of WIDER FACE, achieving new state-of-the-art performance on these two competitive benchmarks. Code is available at https://github.com/zehuichen123/DDOD. CCS CONCEPTS• Computing methodologies → Object detection.

BackgroundHuman papillomavirus (HPV) is one of the most common sexually transmitted viruses. Data about HPV infection in Guizhou is limited.Methods56,768 cervical samples were collected and genotyped for 15 main high risk and 6 main low risk HPV types.Results16.95% (9623/56768) of samples were HPV positive; 90.70% (8728/9623) of HPV positive women were infected by high risk HPV. High risk and high risk mix infection (1458; 70.85%) was the most common mix HPV infection type. The highest HPV detection rate was found in age group 41–45 years old (detection rate = 17.89%) (χ2 = 204.77; P < 0.001); the highest within-group HPV infection rates were found in the ≤20 (25.62%) and ≥ 61 (24.67%) years old age groups, the lowest within-group HPV infection rate was found in the 31–35 years old age group (15.02%). The highest mix infection proportions were found in the ≥61 (36.06%) and ≤ 20 (33.63%) years old age groups (χ2 = 111.21; P < 0.001), the lowest mix infection proportion was found in the 41–45 (17.42%) years old age group. The highest high risk infection proportions were found in the 26–30 (92.98%), ≥61 (92.68%), and 36–40 (92.16%) years old age groups (χ2 = 31.72; P < 0.001), the lowest high risk infection proportion was found in the ≤20 (84.96%) years old age group. HPV infection rates varied with seasons in Guizhou.ConclusionsCharacteristics of HPV distribution in Guizhou were identified. There were significant differences in HPV distribution among age groups, prevention strategies should be adjusted according to the characteristics.

This paper aims to address the problem of supervised monocular depth estimation. We start with a meticulous pilot study to demonstrate that the long-range correlation is essential for accurate depth estimation. Therefore, we propose to leverage the Transformer to model this global context with an effective attention mechanism. We also adopt an additional convolution branch to preserve the local information as the Transformer lacks the spatial inductive bias in modeling such contents. However, independent branches lead to a shortage of connections between features. To bridge this gap, we design a hierarchical aggregation and heterogeneous interaction module to enhance the Transformer features via element-wise interaction and model the affinity between the Transformer and the CNN features in a set-to-set translation manner. Due to the unbearable memory cost caused by global attention on high-resolution feature maps, we introduce the deformable scheme to reduce the complexity. Extensive experiments on the KITTI, NYU, and SUN RGB-D datasets demonstrate that our proposed model, termed DepthFormer, surpasses state-of-the-art monocular depth estimation methods with prominent margins. Notably, it achieves the most competitive result on the highly competitive KITTI depth estimation benchmark. Our codes and models are available 1 .

No abstract