Little is known about the changes in soil microbial phosphorus (P) cycling potential during terrestrial ecosystem management and restoration, although much research aims to enhance soil P cycling. Here, we used metagenomic sequencing to analyse 18 soil microbial communities at a P-deficient degraded mine site in southern China where ecological restoration was implemented using two soil ameliorants and eight plant species. Our results show that the relative abundances of key genes governing soil microbial P-cycling potential were higher at the restored site than at the unrestored site, indicating enhancement of soil P cycling following restoration. The gcd gene, encoding an enzyme that mediates inorganic P solubilization, was predominant across soil samples and was a major determinant of bioavailable soil P. We reconstructed 39 near-complete bacterial genomes harboring gcd, which represented diverse novel phosphate-solubilizing microbial taxa. Strong correlations were found between the relative abundance of these genomes and bioavailable soil P, suggesting their contributions to the enhancement of soil P cycling. Moreover, 84 mobile genetic elements were detected in the scaffolds containing gcd in the 39 genomes, providing evidence for the role of phage-related horizontal gene transfer in assisting soil microbes to acquire new metabolic potential related to P cycling.

No abstract

No abstract

In this paper, we propose a novel query design for the transformer-based object detection. In previous transformer-based detectors, the object queries are a set of learned embeddings. However, each learned embedding does not have an explicit physical meaning and we cannot explain where it will focus on. It is difficult to optimize as the prediction slot of each object query does not have a specific mode. In other words, each object query will not focus on a specific region. To solve these problems, in our query design, object queries are based on anchor points, which are widely used in CNN-based detectors. So each object query focuses on the objects near the anchor point. Moreover, our query design can predict multiple objects at one position to solve the difficulty: ``one region, multiple objects''. In addition, we design an attention variant, which can reduce the memory cost while achieving similar or better performance than the standard attention in DETR. Thanks to the query design and the attention variant, the proposed detector that we called Anchor DETR, can achieve better performance and run faster than the DETR with 10x fewer training epochs. For example, it achieves 44.2 AP with 19 FPS on the MSCOCO dataset when using the ResNet50-DC5 feature for training 50 epochs. Extensive experiments on the MSCOCO benchmark prove the effectiveness of the proposed methods. Code is available at https://github.com/megvii-research/AnchorDETR.

Object detectors are usually equipped with backbone networks designed for image classification. It might be sub-optimal because of the gap between the tasks of image classification and object detection. In this work, we present DetNAS to use Neural Architecture Search (NAS) for the design of better backbones for object detection. It is non-trivial because detection training typically needs ImageNet pre-training while NAS systems require accuracies on the target detection task as supervisory signals. Based on the technique of one-shot supernet, which contains all possible networks in the search space, we propose a framework for backbone search on object detection. We train the supernet under the typical detector training schedule: ImageNet pre-training and detection fine-tuning. Then, the architecture search is performed on the trained supernet, using the detection task as the guidance. This framework makes NAS on backbones very efficient. In experiments, we show the effectiveness of DetNAS on various detectors, for instance, one-stage RetinaNet and the two-stage FPN. We empirically find that networks searched on object detection shows consistent superiority compared to those searched on ImageNet classification. The resulting architecture achieves superior performance than hand-crafted networks on COCO with much less FLOPs complexity. Code and models have been made available at: https://github.com/megvii-model/DetNAS.