HRYNet: A Highly Robust YOLO Network for Complex Road Traffic Object Detection.

Object detection is a crucial component of the perception system in autonomous driving. However, the road scene presents a highly intricate environment where the visibility and characteristics of traffic targets are susceptible to attenuation and loss due to various complex road scenarios such as lighting conditions, weather conditions, time of day, background elements, and traffic density. Nevertheless, the current object detection network must exhibit more learning capabilities when detecting such targets. This also exacerbates the loss of features during the feature extraction and fusion process, significantly compromising the network's detection performance on traffic targets. This paper presents a novel methodology by which to overcome the concerns above, namely HRYNet. Firstly, a dual fusion gradual pyramid structure (DFGPN) is introduced, which employs a two-stage gradient fusion strategy to enhance the generation of more comprehensive multi-scale high-level semantic information, strengthen the interconnection between non-adjacent feature layers, and reduce the information gap that exists between them. HRYNet introduces an anti-interference feature extraction module, the residual multi-head self-attention mechanism (RMA). RMA enhances the target information by implementing a characteristic channel weighting policy, thereby reducing background interference and improving the attention capability of the network. Finally, the detection performance of HRYNet was evaluated by utilizing three datasets: the horizontally collected dataset BDD1000K, the UAV high-altitude dataset Visdrone, and a custom dataset. Experimental results demonstrate that HRYNet achieves a higher mAP_0.5 compared with YOLOv8s on the three datasets, with increases of 10.8%, 16.7%, and 5.5%, respectively. To optimize HRYNet for mobile devices, this study presents Lightweight HRYNet (LHRYNet), which effectively reduces the number of model parameters by 2 million. The results demonstrate that LHRYNet outperforms YOLOv8s in terms of mAP_0.5, with improvements of 6.7%, 10.9%, and 2.5% observed on the three datasets, respectively.

Divergent molecular events underlying initial T-cell commitment in human prenatal and postnatal thymus.

Introduction: Intrathymic T-cell development is a coordinated process accompanied by dynamic changes in gene expression. Although the transcriptome characteristics of developing T cells in both human fetal and postnatal thymus at single-cell resolution have been revealed recently, the differences between human prenatal and postnatal thymocytes regarding the ontogeny and early events of T-cell development still remain obscure. Moreover, the transcriptional heterogeneity and posttranscriptional gene expression regulation such as alternative polyadenylation at different stages are also unknown. Method: In this study, we performed integrative single-cell analyses of thymocytes at distinct developmental stages. Results: The subsets of prenatal CD4-CD8- double-negative (DN) cells, the most immature thymocytes responsible for T-cell lineage commitment, were characterized. By comprehensively comparing prenatal and postnatal DN cells, we revealed significant differences in some key gene expressions. Specifically, prenatal DN subpopulations exhibited distinct biological processes and markedly activated several metabolic programs that may be coordinated to meet the required bioenergetic demands. Although showing similar gene expression patterns along the developmental path, prenatal and postnatal thymocytes were remarkably varied regarding the expression dynamics of some pivotal genes for cell cycle, metabolism, signaling pathway, thymus homing, and T-cell commitment. Finally, we quantified the transcriptome-wide changes in alternative polyadenylation across T-cell development and found diverse preferences of polyadenylation site usage in divergent populations along the T-cell commitment trajectory. Discussion: In summary, our results revealed transcriptional heterogeneity and a dynamic landscape of alternative polyadenylation during T-cell development in both human prenatal and postnatal thymus, providing a comprehensive resource for understanding T lymphopoiesis in human thymus.

Network pharmacology and bioinformatics were used to construct a prognostic model and immunoassay of core target genes in the combination of quercetin and kaempferol in the treatment of colorectal cancer.

Purpose: CRC is a malignant tumor seriously threatening human health. Quercetin and kaempferol are representative components of traditional Chinese medicine (TCM). Previous studies have shown that both quercetin and kaempferol have antitumor pharmacological effects, nevertheless, the underlying mechanism of action remains unclear. To explore the synergy and mechanism of quercetin and kaempferol in colorectal cancer. Methods: In this study, network pharmacology, and bioinformatics are used to obtain the intersection of drug targets and disease genes. Training gene sets were acquired from the TCGA database, acquired prognostic-related genes by univariate Cox, multivariate Cox, and Lasso-Cox regression models, and validated in the GEO dataset. We also made predictions of the immune function of the samples and used molecular docking to map a model for binding two components to prognostic genes. Results: Through Lasso-Cox regression analysis, we obtained three models of drug target genes. This model predicts the combined role of quercetin and kaempferol in the treatment and prognosis of CRC. Prognostic genes are correlated with immune checkpoints and immune infiltration and play an adjuvant role in the immunotherapy of CRC. Conclusion: Core genes are regulated by quercetin and kaempferol to improve the patient's immune system and thus assist in the treatment of CRC.

Single-cell transcriptomics uncover the key ferroptosis regulators contribute to cancer progression in head and neck squamous cell carcinoma.

The mechanism underlying the association between the development of head and neck squamous cell carcinoma (HNSCC) and ferroptosis is unclear. We analyzed the transcriptomes of 5902 single cells from a single-cell RNA-sequencing (scRNA-seq) dataset. They then aggregate into B cells, epithelial cells, fibroblasts, germ cells, mesenchymal cells, cancer stem cells, stem cells, T cells and endometrial cells, respectively. Our study shows that multiple pathways are significantly enriched in HNSCC development including extracellular matrix structural components, humoral immune responses, and muscle contraction. Differentially expressed genes analysis in Pseudotime analysis, pathway and biological function indicated that there was a significant correlation in the ferroptosis pathway. Furthermore, higher ferroptosis potential index (FPI) scores were significantly associated with worse overall survival prognosis in HNSCC patients. Pseudo-temporal, survival analyses and immunohistochemistry identified multiple central genes in HNSCC development, including ACSL1, SLC39A14, TFRC, and PRNP genes, and indicated associated ferroptosis. Overall, our study detected ferroptosis-related features is closely correlated with HNSCC prognosis and development, and deserved candidates suitable for immunotherapy treatment strategies determination for HNSCC patients.