4545 papers
SpikeTrainPatternDetector (STPD) is an open-source R/Shiny framework designed to preserve the provenance of neuronal firing-pattern detection decisions through automated candidate generation, deterministic conflict resolution, and manual review, serving as a reproducible infrastructure for expert validation rather than a standalone biological classifier.
This study demonstrates that in tropical coastal restinga forests, trees and shrubs adjust their root water-uptake depth in response to water table levels to maintain water status, but this belowground investment creates a trade-off that specifically constrains aboveground photosynthetic traits and greenness in trees.
This study demonstrates that sublethal exposure to the organophosphate insecticide chlorpyrifos induces significant metabolic stress, biochemical imbalances in carbohydrate and lipid profiles, and severe multi-organ histopathological damage in the freshwater fish *Puntius chola*, highlighting its potential as a sensitive biomarker for monitoring pesticide contamination in aquatic ecosystems.
This paper introduces Streamline-Based Analysis (SBA), a novel non-parametric framework that performs statistical inference directly on individual tractography streamlines to achieve high-resolution, whole-brain white matter analysis with family-wise error control, thereby bridging the gap between voxel-, fixel-, and tract-level methods.
This study identifies GDI2 as a critical inhibitor of adipocyte lipophagy that suppresses beige fat thermogenesis and promotes diet-induced obesity, demonstrating that its deletion enhances lipid degradation and metabolic health by activating a PPARα/PGC1α-dependent transcriptional program.
This paper demonstrates that a z-averaged ensemble of three representationally orthogonal deep-learning models significantly improves protein–ligand virtual screening performance on leakage-controlled benchmarks by achieving genuine out-of-distribution generalization, thereby overcoming the limitations of single-model approaches on novel targets.
This paper presents a reproducible, end-to-end workflow for selecting, optimizing, and deploying accurate, auditable bioacoustics models on CPU-only edge devices, utilizing a specific architecture of log-mel spectrograms, vision-pretrained CNNs, and attention mechanisms to overcome real-world challenges like noise and domain shift while validated on the BirdCLEF+ 2026 Pantanal benchmark.
The bacterium *Rhodococcus ruber* SiAl demonstrates highly selective and efficient removal of iron (up to 99%) and phosphate from culture media, a capability supported by identified siderophore and phosphate transport genes, while showing minimal magnesium uptake and growth inhibition by manganese.
This study developed a simple and practical cultivar identification system for Japanese pummelo using 18 selected InDel markers detectable via standard PCR and gel electrophoresis, which not only enables efficient protection of breeders' rights but also provides insights into the genetic relationships and parentage of key cultivars.
The paper introduces AIGCRS-AMP30, an integrated AI framework that combines diffusion-based generation, classification, and regression models to design short, potent antimicrobial peptides, which were experimentally validated to show high efficacy against multidrug-resistant pathogens with low host toxicity.