Explore the vast range of titles available.

Understanding the functions of proteins is of great importance for deciphering the mechanisms of life activities. To date, there have been over 200 million known proteins, but only 0.2% of them have well-annotated functional terms. By measuring the contacts among residues, proteins can be described as graphs so that the graph leaning approaches can be applied to learn protein representations. However, existing graph-based methods put efforts in enriching the residue node information and did not fully exploit the edge information, which leads to suboptimal representations considering the strong association of residue contacts to protein structures and to the functions. In this article, we propose SuperEdgeGO, which introduces the supervision of edges in protein graphs to learn a better graph representation for protein function prediction. Different from common graph convolution methods that uses edge information in a plain or unsupervised way, we introduce a supervised attention to encode the residue contacts explicitly into the protein representation. Comprehensive experiments demonstrate that SuperEdgeGO achieves state-of-the-art performance on all three categories of protein functions. Additional ablation analysis further proves the effectiveness of the devised edge supervision strategy. The implementation of edge supervision in SuperEdgeGO resulted in enhanced graph representations for protein function prediction, as demonstrated by its superior performance across all the evaluated categories. This superior performance was confirmed through ablation analysis, which validated the effectiveness of the edge supervision strategy. This strategy has a broad application prospect in the study of protein function and related fields.

Understanding the function of proteins is of great significance for revealing disease pathogenesis and discovering new targets. Benefiting from the explosive growth of the protein universal, deep learning has been applied to accelerate the protein annotation cycle from different biological modalities. However, most existing deep learning-based methods not only fail to effectively fuse different biological modalities, resulting in low-quality protein representations, but also suffer from the convergence of suboptimal solution caused by sparse label representations. Aiming at the above issue, we propose a multiprocedural approach for fusing heterogeneous biological modalities and annotating protein functions, i.e., MIF2GO (Multimodal Information Fusion to infer Gene Ontology terms), which sequentially fuses up to six biological modalities ranging from different biological levels in three steps, thus leading to powerful protein representations. Evaluation results on seven benchmark datasets show that the proposed method not only considerably outperforms state-of-the-art performance, but also demonstrates great robustness and generalizability across species. Besides, we also present biological insights into the associations between those modalities and protein functions. This research provides a robust framework for integrating multimodal biological data, offering a scalable solution for protein function annotation, ultimately facilitating advancements in precision medicine and the discovery of novel therapeutic strategies. MIF2GO leverages up to six biological modalities to enhance protein function annotation. It outperforms state-of-the-art methods, showing robustness and generalizability across species, while offering insights into modality-function associations.

The substantial domain discrepancy inherent in multi-source and multi-view imagery presents formidable challenges to achieving precise drone-based multi-view geo-localization. Existing methodologies primarily focus on designing sophisticated backbone architectures to extract view-invariant representations within abstract feature spaces, yet they often overlook the rich and discriminative frequency-domain cues embedded in multi-view data. Inspired by the principles of π-Noise theory, this paper proposes a frequency-domain Positive-Incentive Information Screening (PIIS) mechanism that adaptively identifies and preserves task-relevant frequency components based on entropy-guided information metrics. This principled approach selectively enhances discriminative spectral signatures while suppressing redundant or noisy components, thereby improving multi-view feature alignment under substantial appearance and geometric variations. The proposed PIIS strategy demonstrates strong architectural generality, as it can be seamlessly integrated into various backbone networks including convolutional-based and Transformer-based architectures while maintaining consistent performance improvements across different models. Extensive evaluations on the University-1652 and SUES-200 datasets have validated the great potential of the proposed method. Specifically, the PIIS-N model achieves a Recall@1 of 94.56% and a mean Average Precision (mAP) of 95.44% on the University-1652 dataset, exhibiting competitive accuracy among contemporary approaches. These findings underscore the considerable promise of frequency-domain analysis in advancing multi-view geo-localization.

During molecular surveillance of human adenoviruses (HAdVs) in children hospitalized with acute lower respiratory tract infections in Beijing, China, during 2014-2024, the most prevalent genotypes were HAdV-B114 (53.85%) and HAdV-B7 (27.18%). A novel recombinant genotype, HAdV-B117, was identified in 2 children <5 years of age with severe community-acquired pneumonia and serious complications. Genomic analysis revealed that HAdV-B117 arose from HAdV-B114 (P7H3F3) with the fiber gene from HAdV-B7. We observed amino acid substitutions and deletions in the pivotal regions of 3 major capsid proteins, and some were predicted to alter the protein structure. In vitro, the replication kinetics of HAdV-B117 were similar to those of HAdV-B3 and HAdV-B7. Clinical manifestations resembled severe pneumonia caused by HAdV-B3 or HAdV-B7. Both children recovered after treatment. The emergence of HAdV-B117 highlights the need for continuous genomic surveillance of HAdVs to detect novel recombinants with potential public health effects.

Whether monitoring of the methotrexate (MTX) concentrations after high-dose MTX (HD-MTX) infusion can predict toxicities is still controversial, especially when HD-MTX therapy is used in the treatment of children with acute lymphoblastic leukemia (ALL), which is different than the previous schedules. The relationship between patient characteristics and severe adverse events (AEs) has yet to be determined. To analyze the relationship between the MTX concentration and toxicities and to identify the risk predictors from patient characteristics for severe AEs during HD-MTX therapy in children with ALL. We conducted a retrospective study on children with ALL who were treated with 388 HD-MTX infusions. The chi-square test and univariate and logistic regression analyses were used to analyze the relationship between the MTX concentrations and toxicities and to identify predictors for severe AEs. Febrile neutropenia ( =0.000) and vomiting ( =0.034) were more likely to occur if the infusion had an MTX level ≥1 μmol/L at 44 h, but other toxicities had no correlations with MTX concentration. Predictive factors for toxicities were as follows: higher risk stratification and higher values of albumin (ALB) for leucopenia, higher values of white blood cell count (WBC) for anemia, higher values of ALB and creatinine (Cr) for neutropenia, higher risk stratification and higher 44-h MTX concentration for febrile neutropenia, higher values of alanine transferase (ALT) for elevated ALT, higher values of ALT for elevated aspartate transferase (AST), and higher values of total bilirubin (TBil) for vomiting. Routine monitoring of 44-h MTX concentrations is essential to identify patients at high risk of developing febrile neutropenia and vomiting. This study may provide a reference for clinicians to distinguish patients with a relatively high risk of severe AEs based on certain characteristics before HD-MTX infusion.

Background Porcine reproductive and respiratory syndrome virus (PRRSV) is an enveloped virus, bearing severe economic consequences to the swine industry worldwide. Previous studies on enveloped viruses have shown that many incorporated cellular proteins associated with the virion's membranes that might play important roles in viral infectivity. In this study, we sought to proteomically profile the cellular proteins incorporated into or associated with the virions of a highly virulent PRRSV strain GDBY1, and to provide foundation for further investigations on the roles of incorporated/associated cellular proteins on PRRSV's infectivity. Results In our experiment, sixty one cellular proteins were identified in highly purified PRRSV virions by two-dimensional gel electrophoresis coupled with mass spectrometric approaches. The identified cellular proteins could be grouped into eight functional categories including cytoskeletal proteins, chaperones, macromolecular biosynthesis proteins, metabolism-associated proteins, calcium-dependent membrane-binding proteins and other functional proteins. Among the identified proteins, four have not yet been reported in other studied envelope viruses, namely, guanine nucleotide-binding proteins, tyrosine 3-monooxygenase/tryptophan 5-monooxygenase, peroxiredoxin 1 and galectin-1 protein. The presence of five selected cellular proteins (i.e., β-actin, Tubulin, Annexin A2, heat shock protein Hsp27, and calcium binding proteins S100) in the highly purified PRRSV virions was validated by Western blot and immunogold labeling assays. Conclusions Taken together, the present study has demonstrated the incorporation of cellular proteins in PRRSV virions, which provides valuable information for the further investigations for the effects of individual cellular proteins on the viral replication, assembly, and pathogenesis.

This study aimed to construct a predictive model for recurrence and metastasis in patients with localized clear cell renal cell carcinoma (ccRCC) based on multiple preoperative blood indexes and oncological characteristics. Overall, 442 patients with localized ccRCC between 2013 and 2015 were included. Using least absolute shrinkage and selection operator (LASSO) Cox regression analysis, the top three risk factors from the peripheral blood indicators were screened to construct a risk score, and a prognostic model was established. Harrell's concordance index (C-index) was applied to evaluate the predictive accuracy of the model for predicting disease-free survival (DFS) in ccRCC. Out of 38 blood indexes, the top three predictors were fibrinogen (FIB), C-reactive protein (CRP) and neutrophil-lymphocyte ratio (NLR). The FIB-CRP-NLR (FCN) score (hazard ratio [HR]: 1.86, 95% confidence interval [CI]: 1.21-2.9, = 0.005) was an independent prognostic factor in multivariate analysis. Furthermore, the FIB-CRP-NLR-T-Grade (FCNTG) risk model combining FCN score, T stage and Furhman grade achieved a higher prognostic accuracy (mean C-index, 0.728) than both the FCN score alone (mean C-index, 0.675) and the stage, size, grade, and necrosis (SSIGN) score (mean C-index, 0.686) in the validation cohort. The FCN score combining peripheral blood indicators of inflammation and coagulation is an independent prognostic marker of ccRCC. The FCNTG model, which systemically incorporates preoperative blood indexes to oncological characteristics, shows its advantages of convenience and high prediction efficiency.

Background Newcastle disease virus (NDV) is an enveloped RNA virus, bearing severe economic losses to the poultry industry worldwide. Previous virion proteomic studies have shown that enveloped viruses carry multiple host cellular proteins both internally and externally during their life cycle. To address whether it also occurred during NDV infection, we performed a comprehensive proteomic analysis of highly purified NDV La Sota strain particles. Results In addition to five viral structural proteins, we detected thirty cellular proteins associated with purified NDV La Sota particles. The identified cellular proteins comprised several functional categories, including cytoskeleton proteins, annexins, molecular chaperones, chromatin modifying proteins, enzymes-binding proteins, calcium-binding proteins and signal transduction-associated proteins. Among these, three host proteins have not been previously reported in virions of other virus families, including two signal transduction-associated proteins (syntenin and Ras small GTPase) and one tumor-associated protein (tumor protein D52). The presence of five selected cellular proteins (i.e., β-actin, tubulin, annexin A2, heat shock protein Hsp90 and ezrin) associated with the purified NDV particles was validated by Western blot or immunogold labeling assays. Conclusions The current study presented the first standard proteomic profile of NDV. The results demonstrated the incorporation of cellular proteins in NDV particles, which provides valuable information for elucidating viral infection and pathogenesis.

Background Infectious bronchitis virus (IBV) is the coronavirus of domestic chickens causing major economic losses to the poultry industry. Because of the complexity of the IBV life cycle and the small number of viral structural proteins, important virus-host relationships likely remain to be discovered. Toward this goal, we performed two-dimensional gel electrophoresis fractionation coupled to mass spectrometry identification approaches to perform a comprehensive proteomic analysis of purified IBV particles. Results Apart from the virus-encoded structural proteins, we detected 60 host proteins in the purified virions which can be grouped into several functional categories including intracellular trafficking proteins (20%), molecular chaperone (18%), macromolcular biosynthesis proteins (17%), cytoskeletal proteins (15%), signal transport proteins (15%), protein degradation (8%), chromosome associated proteins (2%), ribosomal proteins (2%), and other function proteins (3%). Interestingly, 21 of the total host proteins have not been reported to be present in virions of other virus families, such as major vault protein, TENP protein, ovalbumin, and scavenger receptor protein. Following identification of the host proteins by proteomic methods, the presence of 4 proteins in the purified IBV preparation was verified by western blotting and immunogold labeling detection. Conclusions The results present the first standard proteomic profile of IBV and may facilitate the understanding of the pathogenic mechanisms.

Heatmap-based anatomical landmark detection is still facing two unresolved challenges: 1) inability to accurately evaluate the distribution of heatmap; 2) inability to effectively exploit global spatial structure information. To address the computational inability challenge, we propose a novel position-aware and sample-aware central loss. Specifically, our central loss can absorb position information, enabling accurate evaluation of the heatmap distribution. More advanced is that our central loss is sample-aware, which can adaptively distinguish easy and hard samples and make the model more focused on hard samples while solving the challenge of extreme imbalance between landmarks and non-landmarks. To address the challenge of ignoring structure information, a Coordinated Transformer, called CoorTransformer, is proposed, which establishes long-range dependencies under the guidance of landmark coordination information, making the attention more focused on the sparse landmarks while taking advantage of global spatial structure. Furthermore, CoorTransformer can speed up convergence, effectively avoiding the defect that Transformers have difficulty converging in sparse representation learning. Using the advanced CoorTransformer and central loss, we propose a generalized detection model that can handle various scenarios, inherently exploiting the underlying relationship between landmarks and incorporating rich structural knowledge around the target landmarks. We analyzed and evaluated CoorTransformer and central loss on three challenging landmark detection tasks. The experimental results show that our CoorTransformer outperforms state-of-the-art methods, and the central loss significantly improves the performance of the model with p-values< 0.05.