4404 papers
This paper presents a reproducible, CPU-accessible machine-learning framework that employs multi-objective active learning to navigate the vast LNP formulation space, successfully identifying lipid compositions that maximize brain delivery while minimizing liver tropism, thereby overcoming a major barrier to CNS mRNA therapeutics.
This study demonstrates that tropical forest trees exhibit modest, species-specific plasticity in photosystem II heat tolerance during a non-stress seasonal warming period, revealing that while community-level acclimation occurs, certain species already operate near functional thermal limits regardless of their leaf habit or successional status.
This study identifies DACH1 as a novel candidate epigenetic regulator in cervical squamous cell carcinoma, demonstrating through network analysis and TCGA data that its expression is inversely associated with promoter methylation of the tumor-suppressor genes CADM1, ESR1, and SLIT2, independent of global tumor hypermethylation.
This study highlights that the invasive boxelder tree (*Acer negundo*) possesses persistent, photosynthetically active chlorenchyma on its branches that utilizes high salicylic acid levels and efficient non-photochemical quenching mechanisms to protect its photosynthetic apparatus from light stress while maintaining activity throughout the year.
This study demonstrates that dried menstrual blood collected via Q-Pads and stored at room temperature for over two years is a viable, minimally invasive biospecimen for genome-wide DNA methylation profiling and accurate epigenetic age estimation.
This study reveals that archaeal transcriptional regulation is far more diverse than previously thought by identifying 250 non-canonical DNA-binding protein families across a phylogenetically balanced genome panel, thereby expanding the known regulatory landscape beyond the nineteen canonical families and providing a ranked set of experimentally testable hypotheses for their functions.
This study demonstrates that a brief 10-minute mindfulness induction enhances emotional attentional regulation by reducing processing delays for negative stimuli and increasing attention allocation to positive distractors within the Emotional Attentional Blink paradigm.
This study presents an end-to-end deep learning framework for automated mosquito egg detection and counting in vector surveillance that achieves high precision and reliable performance across diverse real-world conditions by leveraging a standardized scanning protocol, expert-guided annotation, and a synthetic data generation pipeline trained exclusively on synthetic images.
This study employs computational in silico screening to identify 11-O-Galloyl bergenin and related Himalayan phytochemicals as novel, high-affinity inhibitors of the oncogenic chaperone AGR2, establishing a preliminary structure-activity relationship for future experimental validation in cancer therapeutics.
This paper proposes the Resonance of Closed Neural Network Geometries (RCNNG) framework, which posits that perception and consciousness emerge from stable resonant attractors in recurrent neural circuits, where perceptual stability and subjective experience ("innergence") arise from the mapping of input-driven closed neural geometries to their intrinsic resonant identities rather than their physical structures.