Search

Effective prevention strategies for post-COVID complications are crucial for patients, clinicians, and policy makers to mitigate their cumulative burden. This study evaluated the association of modifiable lifestyle factors (smoking, alcohol intake, BMI, physical activity, sedentary time, sleep duration, and dietary habits) with COVID-19 multisystem sequelae, death, and hospitalization in the UK Biobank cohort (n = 68,896). A favorable lifestyle (6-10 healthy factors; 46.4%) was associated with a 36% lower risk of multisystem sequelae (HR, 0.64; 95% CI, 0.58-0.69; ARR at 210 days, 7.08%; 95% CI, 5.98-8.09) compared to an unfavorable lifestyle (0-4 factors; 12.3%). Risk reductions spanned all 10 organ systems, including cardiovascular, coagulation, metabolic, gastrointestinal, kidney, mental health, musculoskeletal, respiratory disorders, and fatigue. This beneficial effect was largely attributable to direct lifestyle impacts independent of corresponding pre-infection comorbidities (71% for any sequelae). A favorable lifestyle was also related to the risk of post-COVID death (HR 0.59, 0.52-0.66) and hospitalization (HR 0.78, 0.73-0.84). These associations persisted across acute and post-acute infection phases, irrespective of hospitalization status, vaccination, or SARS-CoV-2 variant. These findings underscore the clinical and public health importance of adhering to a healthy lifestyle in mitigating long-term COVID-19 adverse impacts and enhancing future pandemic preparedness. The impact of modifiable lifestyle factors on the risk of post-COVID-19 sequelae is not well understood. Here, the authors perform a prospective cohort study amongst UK Biobank participants and show that a healthy lifestyle is associated with lower risk of post-COVID-19 sequelae across multiple organ systems. [ABSTRACT FROM AUTHOR]

Road extraction from remote-sensing images is of great significance for vehicle navigation and emergency insurance. However, the road information extracted in the remote-sensing image is discontinuous because the road in the image is often obscured by the shadows of trees or buildings. Moreover, due to the scale difference of roads in remote-sensing images, it remains a computational challenge to extract small-size roads from remote-sensing images. To address those problems, we propose a road extraction method based on adaptive global feature fusion (AGF-Net). First, a dilated convolution strip attention (DCSA) module is designed from the encoder–decoder structure. It consists of the dilated convolution and the strip attention module, which adaptively emphasizes relevant features in vertical and horizontal directions. Then, multiple global feature fusion modules (GFFM) in the skip connection are designed to supplement the decoder with road detail features, and we design a multi-scale strip convolution module (MSCM) to implement the GFFM module to obtain multi-scale road information. We compare AGF-Net to state-of-the-art methods and report their performance using standard evaluation metrics, including Intersection over Union (IoU), F1-score, precision, and recall. Our proposed AGF-Net achieves higher accuracy compared to other existing methods on the Massachusetts Road Dataset, DeepGlobe Road Dataset, CHN6-CUG Road Dataset, and BJRoad Dataset. The IoU obtained on these datasets are 0.679, 0.673, 0.567, and 0.637, respectively. [ABSTRACT FROM AUTHOR]

RNA-binding proteins play crucial roles in the regulation of gene expression, and understanding the interactions between RNAs and RBPs in distinct cellular conditions forms the basis for comprehending the underlying RNA function. However, current computational methods pose challenges to the cross-prediction of RNA-protein binding events across diverse cell lines and tissue contexts. Here, we develop HDRNet, an end-to-end deep learning-based framework to precisely predict dynamic RBP binding events under diverse cellular conditions. Our results demonstrate that HDRNet can accurately and efficiently identify binding sites, particularly for dynamic prediction, outperforming other state-of-the-art models on 261 linear RNA datasets from both eCLIP and CLIP-seq, supplemented with additional tissue data. Moreover, we conduct motif and interpretation analyses to provide fresh insights into the pathological mechanisms underlying RNA-RBP interactions from various perspectives. Our functional genomic analysis further explores the gene-human disease associations, uncovering previously uncharacterized observations for a broad range of genetic disorders. Predicting dynamic RNA-RBP interactions in diverse cell lines is an important challenge in unravelling RNA function and post-transcriptional regulatory mechanisms. Here, authors develop HDRNet, an end-to-end deep-learning-based framework for accurately predicting dynamic RBP binding events across various cellular conditions. [ABSTRACT FROM AUTHOR]

Deploying convolutional neural networks (CNNs) on mobile devices is difficult due to the limited memory and computation resources. We aim to design efficient neural networks for heterogeneous devices including CPU and GPU, by exploiting the redundancy in feature maps, which has rarely been investigated in neural architecture design. For CPU-like devices, we propose a novel CPU-efficient Ghost (C-Ghost) module to generate more feature maps from cheap operations. Based on a set of intrinsic feature maps, we apply a series of linear transformations with cheap cost to generate many ghost feature maps that could fully reveal information underlying intrinsic features. The proposed C-Ghost module can be taken as a plug-and-play component to upgrade existing convolutional neural networks. C-Ghost bottlenecks are designed to stack C-Ghost modules, and then the lightweight C-GhostNet can be easily established. We further consider the efficient networks for GPU devices. Without involving too many GPU-inefficient operations (e.g., depth-wise convolution) in a building stage, we propose to utilize the stage-wise feature redundancy to formulate GPU-efficient Ghost (G-Ghost) stage structure. The features in a stage are split into two parts where the first part is processed using the original block with fewer output channels for generating intrinsic features, and the other are generated using cheap operations by exploiting stage-wise redundancy. Experiments conducted on benchmarks demonstrate the effectiveness of the proposed C-Ghost module and the G-Ghost stage. C-GhostNet and G-GhostNet can achieve the optimal trade-off of accuracy and latency for CPU and GPU, respectively. MindSpore code is available at https://gitee.com/mindspore/models/pulls/1809, and PyTorch code is available at https://github.com/huawei-noah/CV-Backbones. [ABSTRACT FROM AUTHOR]

The transition set semantics (Wang and Jiao, LNCS 6128:84-103, ) partitions the Petri net behaviors in a canonical way such that behaviors in an equivalence class have the same canonical transition set sequence. This article extends the semantics in two ways: firstly, the semantics is parameterized by the basic relation on the structural transitions to define different variants; secondly, the semantics for the infinite firing sequences of the net is defined. We prove that these extensions still preserve the well-definedness, soundness and completeness of the semantics. Furthermore, we show how to recognize some infinite sequences called back-loops in the view of this new semantics. [ABSTRACT FROM AUTHOR]

The identification and functional characterization of natural variants in plants are essential for understanding phenotypic adaptation. Here we identify a molecular variation in At2g47310 that contributes to the natural variation in flowering time in Arabidopsis thaliana accessions. This gene, which we term SISTER of FCA (SSF), functions in an antagonistic manner to its close homolog FCA. Genome-wide association analysis screens two major haplotypes of SSF associated with the natural variation in FLC expression, and a single polymorphism, SSF-N414D, is identified as a main contributor. The SSF414N protein variant interacts more strongly with CUL1, a component of the E3 ubiquitination complex, than the SSF414D form, mediating differences in SSF protein degradation and FLC expression. FCA and SSF appear to have arisen through gene duplication after dicot-monocot divergence, with the SSF-N414D polymorphism emerging relatively recently within A. thaliana. This work provides a good example for deciphering the functional importance of natural polymorphisms in different organisms. Natural variation represents valuable source for gene discovery. Here, the authors show that a homolog of Flowering Control Locus A (FCA) functions in an antagonistic manner to FCA in regulating Arabidopsis flowering time through interacting with CUL1-E3 and modulating FLC expression. [ABSTRACT FROM AUTHOR]