Search Results (442)

The grapevine (Vitis vinifera L.) is one of the most important horticultural crops, with thousands of varieties cultivated worldwide. In this study, we analyzed chloroplast SNV markers using a whole-genome shotgun sequencing approach to investigate the genetic diversity and phylogeny of 409 cultivated V. vinifera accessions originating from nine countries across Southeast and Central Europe, as well as a heterogeneous set of additional accessions maintained by INRAE. Shotgun sequencing allowed high coverage, enabling the detection of 93 SNVs across 24 chloroplast genes, including 11 non-synonymous variants. The ycf1 gene showed the highest variability, consistent with its role in species differentiation. Haplotype analysis revealed 102 distinct haplotypes, with clear geographic structuring: ATT predominated in the eastern Mediterranean, ATA in western Europe, and GTA mainly in a heterogeneous group of varieties from a French collection. To validate the shotgun approach, seven SNV markers were analyzed using target capture sequencing, confirming the accuracy of detected variants with only minimal discrepancies, which is mostly attributable to homopolymeric regions and low-frequency alleles. Phylogenetic analyses using both trees and networks delineated three major haplotype clusters, reflecting human-mediated dispersal of grapevine cultivars through historical viticultural practices. This study represents the largest chloroplast genome analysis of cultivated V. vinifera to date, providing a large cpDNA resource for assessing chloroplast diversity and maternal haplotype structure in cultivated grapevine. The results highlight the power of combining high-throughput sequencing and chloroplast genomics for population-level studies in perennial crops. Full article

Coaxial stacking between adjacent stems is a key tertiary interaction that defines the spatial organization of RNA junctions, which are core structural motifs in folded RNAs. The accurate prediction of coaxial stacking is critical for RNA 3D structure modeling, yet existing computational tools remain limited, especially for junctions with variable numbers of branches or complex topologies. Here, we present gCoSRNA, a generalizable computational framework for predicting coaxial stacking configurations using RNA sequence and secondary structure as input. Instead of developing separate models for each junction type, gCoSRNA decomposes multi-way junctions into all possible adjacent stem pairs, termed pseudo two-way junctions, and uses a unified RF classifier to evaluate stacking probabilities. Global stacking configurations are inferred by integrating these pairwise predictions, eliminating the need for explicit junction type classification. Benchmarking on two independent test sets (297 RNA junctions), including CASP15/16 and RNA-Puzzles targets, shows that gCoSRNA achieves consistently high accuracy (mean ~0.89) across junctions with two to seven branches, outperforming existing junction-specific methods. These results highlight the model’s ability to capture higher-order structural features and its potential utility in RNA tertiary structure prediction pipelines. Full article

Background/Objectives: Valve replacement (VR) patients are at high risk of postoperative complications, but reproducible Electronic Health Record (EHR) benchmarks for evaluating sequential AI models in this setting are lacking. We develop a reproducible pipeline that extracts two EHR datasets from MIMIC-IV (a general-purpose and a predictive benchmark dataset) capturing perioperative histories, high-resolution time-series, and clinically motivated outcome labels. Methods: The cohort comprises 3890 VR patients with clinician-guided feature selection across diagnoses, procedures, laboratory measurements, medications, and physiological monitoring. As an exemplary use case, we define ICU readmission at first ICU discharge as a surrogate for postoperative risk and derive a predictive benchmark under strict label-leakage control. We then compare a Transformer model trained on tokenized longitudinal EHR sequences with Transformer and XGBoost baselines trained on aggregated feature statistics, and assess performance differences using paired statistical tests across validation splits. Results: ICU readmission stratified in-hospital and 100-day outcomes, including mortality, complications, and rehospitalization, confirming the clinical relevance of the prediction target. The sequential Transformer achieved 0.87 AUROC and 0.69 AUPRC. Corrected resampled t-tests confirm improved performance over the non-sequential Transformer, while the comparison with XGBoost indicates a favorable trend without conclusive evidence. Conclusions: Our findings suggest that leveraging longitudinal EHR sequences yields higher predictive performance than static feature summaries for postoperative risk prediction. The publicly released preprocessing pipeline and cohort-construction code enable researchers with MIMIC-IV access to reproduce the datasets and provide a robust benchmark for developing and comparing time-series models in post-valve replacement care. Full article

Background/Objectives: Segmental duplications (SDs) are major drivers of genome evolution and structural variation in primates, particularly within acrocentric chromosomes, where rDNA arrays and duplicated sequences are densely clustered. However, the evolutionary dynamics of rDNA-linked SDs across great ape lineages have remained poorly characterized due to longstanding technical limitations in genome assembly. Here, we investigate the organization, copy number variation, and evolutionary conservation of acrocentric SDs in great apes by integrating fluorescence in situ hybridization (FISH) with comparative analyses of telomere-to-telomere (T2T) genome assemblies. Methods: Using eight human-derived fosmid probes targeting SD-enriched regions flanking rDNA arrays, we analyzed multiple individuals from chimpanzee, bonobo, gorilla, and both Bornean and Sumatran orangutans. Results: Our FISH analyses revealed extensive lineage-specific variation in SD copy number and chromosomal distribution, with pronounced heteromorphism in African great apes, particularly gorillas, and more conserved patterns in orangutans. Several SDs showed fixed duplications across species, while others exhibited high levels of polymorphism and individual-specific organization. Conclusions: Comparison with T2T assemblies confirmed consistent genomic localization for a subset of probes, whereas others displayed partial discordance, highlighting the persistent challenges in resolving highly repetitive and structurally dynamic regions even with state-of-the-art assemblies. Genome-wide analyses further revealed species-specific enrichment of SDs on rDNA-bearing chromosomes, with chimpanzees and bonobos showing higher proportions than gorillas, and contrasting patterns between the two orangutan species. Overall, our results demonstrate that rDNA-linked SDs represent highly dynamic genomic compartments that have undergone differential expansion and remodeling during great ape evolution. These regions contribute substantially to inter- and intra-species structural variation and provide a mechanistic substrate for lineage-specific genome evolution, underscoring the importance of integrating cytogenetic and T2T-based approaches to fully capture the complexity of duplicated genomic landscapes. Full article

Unmanned aerial vehicles (UAVs) and birds are typical low-altitude small targets in optical remote sensing, often occupying only a few pixels and exhibiting highly similar appearances, which limits the effectiveness of appearance-based discrimination at long distances and low resolutions. To overcome this, we propose a non-appearance-based classification framework using kinematic and time–frequency features. At the trajectory level, kinematic features—including the coefficient of variation of velocity and acceleration, the Spatiotemporal Box-counting Fractal Dimension (SBFD), and the Local Higuchi Fractal Dimension (LHFD)—quantify multi-scale trajectory complexity. At the scale-variation level, time–frequency features, specifically the Time-Frequency Aware Singular Value Entropy (TF-SVE) derived from bounding-box area sequences, capture non-stationary oscillations from bird wing flapping, reflecting behavioral differences from rigid UAV motion. Experiments on a complex real-world dataset show that stacking these features achieves 99.47% classification accuracy, demonstrating a robust, resolution-invariant, and practically effective approach for non-appearance-based recognition of low-altitude targets. Full article

Considering the nonlinear trends, multi-scale variations, and capacity regeneration phenomena exhibited by battery capacity degradation under real-world conditions, accurately predicting its trajectory remains a critical challenge for ensuring the reliability and safety of electric vehicles. To address this, this study proposes a hybrid prediction framework based on Variational Mode Decomposition and a Transformer–Long Short-Term Memory architecture. Specifically, the proposed Variational Mode Decomposition–Transformer for Time Series–Long Short-Term Memory (VMD–TTS–LSTM) framework first decomposes the capacity sequence using Variational Mode Decomposition. The resulting modal components are then aggregated into high-frequency and low-frequency parts based on their frequency centroids, followed by targeted feature analysis for each part. Subsequently, a simplified Transformer encoder (Transformer for Time Series, TTS) is employed to model high-frequency fluctuations, while a Long Short-Term Memory (LSTM) network captures the long-term degradation trends. Evaluated on charging data from 20 commercial electric vehicles under a long-horizon setting of 20 input steps predicting 100 steps ahead, the proposed method achieves a mean absolute error of 0.9247 and a root mean square error of 1.0151, demonstrating improved accuracy and robustness. The results confirm that the proposed frequency-partitioned, heterogeneous modeling strategy provides a practical and effective solution for battery health prediction and energy management in real-world electric vehicle operation. Full article

Independent vector analysis (IVA) has emerged as a powerful tool for fusing and analyzing functional magnetic resonance imaging (fMRI) data. Applying IVA to multi-task fMRI data enhances analytical power by capturing the relationships across different tasks in order to discover their underlying multivariate relationship to one another. Incorporation of prior information into IVA enhances the separability and interpretability of estimated components. In this paper, we demonstrate successful fusion of multi-task fMRI feature data under two settings: constrained IVA and constrained transposed IVA (tIVA). We show that using these methods for fusing multi-task fMRI feature data offers novel ways to improve the quality and interpretability of the analysis. While constrained IVA extracts components linked to distinct brain networks, tIVA reverses the roles of spatial components and subject profiles, enabling flexible analysis of behavioral effects. We apply both methods to a multi-task fMRI dataset of 247 subjects. We demonstrate that for task-based fMRI, structural MRI (sMRI) references provide a better match for task data than resting-state fMRI (rs-fMRI) references, and using sMRI priors improves identification of group differences in task-related networks, such as the sensory-motor network during the Auditory Oddball (AOD) task. Additionally, constrained tIVA allows for targeted investigation of the effects of behavioral variables by applying them individually during the analysis. For instance, by using the letter number sequence subtest, a measure of working memory, as a behavioral constraint in tIVA, we observed significant group differences in the auditory and sensory-motor networks during the AOD task. Results show that the use of two constrained approaches, guided by well-aligned structural and behavioral references, enables a more comprehensive analysis of underlying brain function as modulated by task. Full article

Drug–target affinity (DTA) prediction is one of the core components of drug discovery. Despite considerable advances in previous research, DTA tasks still face several limitations with insufficient multi-modal information of drugs, the inherent sequence length limitation of protein language models, and single attention mechanisms that fail to capture critical multi-scale features. To alleviate the above limitations, we developed a multi-granularity fusion model for drug–target binding affinity prediction, termed MGF-DTA. This model is composed of three fusion modules, specifically as follows. First, the model extracts deep semantic features of SMILES strings through ChemBERTa-2 and integrates them with molecular fingerprints by using gated fusion to enhance the multi-modal information of drugs. In addition, it employs a residual fusion mechanism to integrate the global embeddings from ESM-2 with the local features obtained by the k-mer and principal component analysis (PCA) method. Finally, a hierarchical attention mechanism is employed to extract multi-granularity features from both drug SMILES strings and protein sequences. Comparative analysis with other mainstream methods on the Davis, KIBA, and BindingDB datasets reveals that the MGF-DTA model exhibits outstanding performance advantages. Further, ablation studies confirm the effectiveness of the model components and case study illustrates its robust generalization capability. Full article

Cardiac arrhythmias, especially premature beats and atrial fibrillation, pose substantial clinical risks and detection hurdles. While deep learning has shown promise for automated arrhythmia diagnosis, single-model architectures often lack sufficient performance in distinguishing these two arrhythmia types. This study seeks to address the limitations of individual deep learning models and boost classification accuracy for premature beats and atrial fibrillation. It proposes an arrhythmia classification model integrating multiscale feature enhancement and bidirectional temporal dependency. First, a four-layer convolutional residual module with skip connections extracts multiscale local electrocardiogram (ECG) features. Then, multi-head self-attention strengthens critical feature global correlations. Next, a bidirectional long-term temporal de-pendency network captures sequence contextual dependencies. Finally, a Dropout-regularized fully connected layer enables six-type arrhythmia classification. Experiments on a fused dataset (MIT-BIH arrhythmia, MIT-BIH atrial fibrillation, and CODE datasets) yield an overall accuracy of 98.55% and F1-score of 0.9531. Notably, the F1-scores for premature beats (0.9916) and atrial fibrillation (0.9888) outperform recent literature by 2.16% and 4.39%, respectively. The model demonstrates robust classification performance with effective identification of the target arrhythmias, highlighting its potential as a supportive tool for automated ECG diagnosis. Full article

To elucidate the potential of integrated multi-omics approaches for studying systemic mechanisms of mycorrhizal fungi in mediating plant-microbe interactions, this study employed the Tuber-inoculated Corylus system as a model to demonstrate how high-throughput profiling can investigate how fungal inoculation reshapes the rhizosphere microbial community and correlates with host metabolism. A pot experiment was conducted comparing inoculated (CTG) and non-inoculated (CK) plants, followed by integrated multi-omics analysis involving high-throughput sequencing (16S/ITS), functional prediction (PICRUSt2/FUNGuild), and metabolomics (UPLC-MS/MS). The results demonstrated that inoculation significantly restructured the fungal community, establishing Tuber as a dominant symbiotic guild and effectively suppressing pathogenic fungi. Although bacterial alpha diversity remained stable, the functional profile shifted markedly toward symbiotic support, including antibiotic biosynthesis and environmental adaptation. Concurrently, root metabolic reprogramming occurred, characterized by upregulation of strigolactones and downregulation of gibberellin A5, suggesting a potential “symbiosis-priority” strategy wherein carbon allocation shifted from structural growth to energy storage, and plant defense transitioned from broad-spectrum resistance to targeted regulation. Multi-omics correlation analysis further revealed notable associations between microbial communities and root metabolites, proposing a model in which Tuber acts as a core regulator that collaborates with the host to assemble a complementary micro-ecosystem. In summary, the integrated approach successfully captured multi-level changes, suggesting that Tuber-Corylus symbiosis constitutes a fungus-driven process that transforms the rhizosphere from a competitive state into a mutualistic state, thereby illustrating the role of mycorrhizal fungi as “ecosystem engineers” and providing a methodological framework for green agriculture research. Full article

Fisheries bycatch, while representing a major ecological concern due to the incidental capture of non-target species, also constitutes an underexplored source of marine biomass with biotechnological potential. This study aimed to generate and characterize bioactive peptides from the muscle tissue of three common bycatch species from the Brazilian coast: Paralonchurus brasiliensis, Micropogonias furnieri, and Hepatus pudibundus. Muscle homogenates were hydrolyzed using either Alcalase or Protamex to produce peptide-rich hydrolysates, which were analyzed through SDS-PAGE, HPLC-UV, MALDI-TOF, and LC-MS/MS. De novo sequencing and bioinformatic analyses predicted bioactivities that were subsequently validated by in vitro assays. The results demonstrated that enzyme selection strongly influenced both peptide profiles and bioactivity. The Protamex hydrolysate of P. brasiliensis (PBP) exhibited potent antifungal activity, inhibiting Candida albicans growth by 81%, whereas the Alcalase hydrolysate (PBA) showed moderate inhibition of Staphylococcus aureus (29%). No significant effect was observed against Escherichia coli. Overall, this study highlights a sustainable strategy for the valorization of fisheries bycatch through the production of bioactive marine peptides and identifies P. brasiliensis hydrolyzed with Protamex as a promising source of anti-Candida peptides for pharmaceutical and nutraceutical applications. Full article

Hemp (Cannabis sativa L.) is one of the oldest cultivated plants in the world. It is a wind-pollinated and heterozygous species, and diverse phenotypes can occur within population varieties. In our study, three different hemp varieties—(‘Carmagnola Selected’ (CS), ‘Tiborszallasi’ (TS) and ‘Finola selection’ (FS))—were grown. Based on visual characteristics, two, five and four phenotypes were identified within CS, TS and FS, respectively. According to Cannabis sativa L. transcriptome data from the Sequence Read Archive (SRA), 4631 single-nucleotide polymorphism (SNP) positions were identified to develop capture probes. DNA was isolated from 171 plants representing selected phenotypes of three cultivars. Next-generation sequencing (NGS) libraries were constructed and hybridized with capture probes for target enrichment. The population structure of the samples was analyzed using SNP data for each genotype. Based on genotype profiles, CS formed a single cluster, while TS and FS were each grouped into two clusters, with phenotypes randomly distributed among them. The GWAS results were visualized using Manhattan plots. Fourteen significant SNPs surpassing the false discovery rate (FDR) of 0.01 were identified for delta-9-tetrahydrocannabinol (delta-9-THC). For cannabigerol (CBG), 12 significant SNPs were detected, and for myrcene, one SNP exceeded the 0.01 FDR threshold. However, plausible genes located 1000 bp to the left and right of the SNP position were identified for all significant SNPs. Full article

Building extraction underpins land-use assessment, urban planning, and disaster mitigation, yet dense urban scenes still cause missed small objects, target adhesion, and ragged contours. We present High-Resolution-Mamba (HR-Mamba), a high-resolution semantic segmentation network that augments a High-Resolution Network (HRNet) parallel backbone with edge-aware and sequence-state modeling. A Canny-enhanced, median-filtered stem stabilizes boundaries under noise; Involution-based residual blocks capture position-specific local geometry; and a Mamba-based State Space Models (Mamba-SSM) global branch captures cross-scale long-range dependencies with linear complexity. Training uses a composite loss of binary cross entropy (BCE), Dice loss, and Boundary loss, with weights selected by joint grid search. We further design a feature-driven adaptive post-processing pipeline that includes geometric feature analysis, multi-strategy simplification, multi-directional regularization, and topological consistency verification to produce regular, smooth, engineering-ready building outlines. On dense urban imagery, HR-Mamba improves F1-score from 80.95% to 83.93%, an absolute increase of 2.98% relative to HRNet. We conclude that HR-Mamba jointly enhances detail fidelity and global consistency and offers a generalizable route for high-resolution building extraction in remote sensing. Full article

Artificial intelligence (AI) is playing an increasingly important role in industrial design, particularly in the aerodynamic optimization of turbine components in aero-engines. This study proposes a turbine blade profile optimization method based on a sequence-to-sequence (Seq2Seq) transformer model. By drawing an analogy between language translation and geometric design generation, the method adopts an encoder–decoder architecture to learn the mapping between blade geometry and its aerodynamic performance. To enhance the interpretability and reliability of model outputs, a performance-matching evaluation framework is introduced. Inspired by similarity metrics in natural language processing, this framework proposes quantifiable indicators to assess the deviation between the predicted aerodynamic performance and the design targets. In a turbine design optimization case, the proposed method successfully generates blade profiles that meet predefined aerodynamic performance requirements, with the optimized design showing a 10.9% reduction in total pressure loss coefficient (from 0.744 to 0.663) and a 0.53% increase in total pressure recovery coefficient (from 0.949 to 0.954), verifying the effectiveness of the Seq2Seq transformer model in capturing design capabilities. It also demonstrates the practical value of performance-matching metrics in evaluating deep learning-assisted design. Taken together, AI-driven optimization approaches hold great promise for aerodynamic design in the energy sector. Full article

Background/Objectives: Klebsiella pneumoniae is a major Gram-negative pathogen associated with community-acquired pneumonia (CAP) and a critical contributor to antimicrobial resistance (AMR). Culture-based diagnostics remain the clinical standard but may underestimate microbial diversity and resistance gene profiles. This pilot study compared pathogen detection and antimicrobial resistance gene (ARG) repertoires in matched K. pneumoniae pure cultures and primary sputum samples using targeted next-generation sequencing (tNGS). Methods: We analyzed 153 sputum samples from patients with CAP. Among 48 culture-positive cases, 22 (14% overall; 54% culture-positive) yielded K. pneumoniae. MALDI-TOF MS, phenotypic drug susceptibility testing, and tNGS were conducted on both culture isolates and matched sputum specimens. Microbial composition, ARG diversity, and method concordance were evaluated, with focused analysis of discordant and fatal cases. Results: K. pneumoniae was detected in 14.4% of all CAP cases and accounted for 54.2% of culture-positive samples. Identification rates differed across methods: 35% by MALDI-TOF MS, 45% by culture tNGS, and 29% by sputum tNGS. Sputum tNGS revealed substantially higher microbial diversity than cultures (3.04 vs. 1.42 species per sample) and detected more than sixfold unique ARGs (38 vs. 7), including clinically relevant determinants that were absent from culture isolates. Concordance was high between MALDI-TOF MS and culture tNGS (κ = 0.712), but low between sputum and culture tNGS (κ = 0.279). Among twelve K. pneumoniae isolates included in AMR analysis, all showed resistance to β-lactams, and two-thirds exhibited MDR/XDR phenotypes. Genotypic screening identified seven ARGs, but major ESBL and carbapenemase genes were not detected, suggesting the presence of alternative resistance mechanisms. Overall, sputum tNGS provided additional etiological and resistome information not captured by cultivation and complemented classical diagnostics in CAP involving K. pneumoniae. Conclusions: Culture-based diagnostics and tNGS provide complementary insights into the detection and resistance profiling of K. pneumoniae in CAP, with sputum tNGS revealing broader microbial and resistome information than pure cultures, while classical methods remain essential for species confirmation and phenotypic AST. An integrated diagnostic approach combining both methodologies may improve pathogen detection, guide antimicrobial therapy, and enhance AMR surveillance in K. pneumoniae-associated CAP. Full article