67639 research outputs found
Sort by
Update on carotenoid and apocarotenoid metabolisms and functions in plants
No full textCarotenoids and their derivative apocarotenoids are diverse isoprenoid metabolites vital to plants and critical to humans. Recent discoveries have expanded our understanding of the intricate mechanisms modulating their metabolism and revealed their new functions in plants. Many new regulators and regulatory modules that potentially link carotenoid metabolism with developmental, hormonal and environmental cues have been unraveled. Emerging evidence also reveals the importance of loss of photosynthetic competence for carotenoid accumulation in chromoplasts. Moreover, apocarotenoids rapidly surface as important regulatory metabolites and signals involved in plant growth and development, stress responses, and communication. In this review, we focus on the latest research in elucidating multifaceted regulatory mechanisms governing carotenoid and apocarotenoid metabolism in plants, provide insights into the differentiation of plastids specialized for carotenoid accumulation, and update on the discoveries and functions of bioactive apocarotenoids. Future research directions to address remaining knowledge gaps are outlined. Collectively, we aim to highlight major advances and exciting discoveries in the field, with the goal of enabling precise and effective augmentation of carotenoids and apocarotenoids with improved growth and stress tolerance in crops.We thank Emalee Wrightstone for creating Figure 1-3, Kawthar Alashoor for Figure 4, and Dr. Mohmaed Salem for valuable discussions. We also thank many colleagues and collaborators for their contributions to the field of research.This work was supported by funding from the Agriculture and Food Research Initiative project (2024-67013-42323) from the U.S. Department of Agriculture’s National Institute of Food and Agriculture and USDA-ARS fund to L.L., from Spanish MCIN/AEI/10.13039/501100011033 and European NextGeneration EU/PRTR programs (grants PID2023-149584NB-I00, PCI2021-121941, RED2022-134577-T and PID2020-115810GB-I00) to M.R-C., and Baseline Funding and Competitive Research Grant (2022) given from King Abdullah University of Science and Technology (KAUST) to S.
4D-Bench: Benchmarking Multi-modal Large Language Models for 4D Object Understanding
No full textMultimodal Large Language Models (MLLMs) have demonstrated impressive 2D image/video understanding capabilities. However, there are no publicly standardized benchmarks to assess the abilities of MLLMs in understanding the 4D objects (3D objects with temporal evolution over time). In this paper, we introduce 4D-Bench, the first benchmark to evaluate the capabilities of MLLMs in 4D object understanding, featuring tasks in 4D object Question Answering (4D object QA) and 4D object captioning. 4D-Bench provides 4D objects with diverse categories, high-quality annotations, and tasks necessitating multi-view spatial-temporal understanding, different from existing 2D image/video-based benchmarks. With 4D-Bench, we evaluate a wide range of open-source and closed-source MLLMs. The results from the 4D object captioning experiment indicate that MLLMs generally exhibit weaker temporal understanding compared to their appearance understanding, notably, while open-source models approach closed-source performance in appearance understanding, they show larger performance gaps in temporal understanding. 4D object QA yields surprising findings: even with simple single-object videos, MLLMs perform poorly, with state-of-the-art GPT-4o achieving only 63\% accuracy compared to the human baseline of 91\%. These findings highlight a substantial gap in 4D object understanding and the need for further advancements in MLLMs
Integrative approaches to understanding and monitoring ecosystem functioning in the Northern Red Sea
No full textCoastal ecosystems are vital for marine biodiversity and provide essential ecosystem services, yet they are increasingly threatened by climate change and anthropogenic pressures. The Northern Red Sea (NRS), particularly the NEOM coastal region, represents an ecologically significant yet understudied region characterized by complex oceanographic conditions and unique biodiversity. In such regions, integrative approaches are essential for understanding and monitoring ecosystem functioning. This is currently not possible due to the lack of continuous high-resolution field measurements. This thesis presents a comprehensive framework for assessing the dynamics of the NEOM marine ecosystem by integrating the available field observations, remote sensing data, numerical modeling, and machine learning techniques. We focus on key ecological components, including benthic habitat mapping, phytoplankton phenology, coastal processes, and enhanced monitoring approaches for chlorophyll-a (Chl-a), a proxy for phytoplankton biomass, estimation. The thesis begins by classifying benthic habitats in the NEOM region using high-resolution multispectral satellite (Sentinel-2) and Uncrewed Aerial Vehicle (UAV) imagery. We employed a random forest classification approach to distinguish geomorphic characteristics (i.e., reef crests, rock or sand flats) and specific benthic habitats (i.e., coral reefs and seagrass meadows). Accuracy assessments are used to validate the classification scheme, demonstrating the potential of combined remote sensing datasets for marine habitat monitoring. This classification reveals extensive coral reef coverage (28% of shallow waters) and seagrass meadows capable of sequestering at least 200 tonnes of carbon annually. Following the classification of benthic habitats, we investigate the seasonal patterns in phytoplankton biomass using satellite-derived Chl-a data, in situ measurements, and hydrodynamic model outputs. We reveal strong links between Chl-a and oceanographic processes such as stratification, water transport, and nutrient availability, and identify key areas with anomalous Chl-a seasonality compared to the general NRS pattern. Specifically, open waters peak in winter due to increased vertical mixing and nutrient availability, while several reef-bound areas exhibit summer peaks linked to localised physical processes. The phytoplankton phenology paradox in one such area, Sharma Lagoon, is then examined in detail using higher-resolution satellite-derived Chl-a data and hydrodynamic model outputs, as well as additional in situ observations. We attribute the lagoon’s opposite Chl-a seasonality to complex interactions between stratification and water exchange. Specifically, its summer peak is driven by tidal oscillations and diurnal heat fluxes that enhance vertical mixing and nutrient availability, contrasting with the stronger winter stratification in the lagoon. Finally, a Bayesian Neural Network (BNN) framework is developed for enhanced Chl-a retrieval, providing uncertainty quantification and addressing limitations of conventional ocean colour algorithms previously encountered. The developed BNN-enhanced model statistically outperforms traditional algorithms by incorporating SST and spatial data, and demonstrates its utility in monitoring coastal development impacts in Sindalah. The implementation of BNNs represents a significant step forward in the operational monitoring of marine ecosystems, offering a reliable tool for detecting ecological shifts. The outcomes of this thesis have direct implications for sustainable coastal management, emphasizing the need for adaptive, data-driven strategies in response to increasing anthropogenic pressures and climate change
Nitrification Inhibition: Uncertainties and Opportunities for Sustainable Agriculture
No full textABSTRACTIn the contemporary discourse on sustainable development, reducing nitrogen (N) pollution is as critical as mitigating carbon dioxide (CO2) emissions. Nitrogen is a vital macronutrient for plant growth, and its application in fertilizers has significantly enhanced crop yields. However, intensive and inefficient N fertilization has led to serious environmental consequences, including water eutrophication, biodiversity loss, and increased emissions of nitrous oxide (N2O), a potent greenhouse gas for which agriculture is the main anthropogenic source. Stabilizing fertilizers with nitrification inhibitors (NIs) presents a promising strategy to improve N use efficiency and mitigate N losses. Despite their demonstrated benefits, NI-stabilized fertilizers still represent only a minor share of the global fertilizer market. This opinion article explores the key uncertainties that may be limiting their broader adoption from an economic and ecological perspective. We examine current knowledge gaps regarding the effects of NIs on soil health and microbial communities, the potential for resistance development among nitrifiers, and the context-dependent variability in their field performance. We also emphasize the need for full life cycle assessments to evaluate whether their environmental benefits outweigh the costs associated with production and application. Finally, we propose strategies to optimize both the use and design of NIs, such as soil-specific application approaches, decoupling NI dosage from N rates, and the discovery of more potent and selective inhibitors. By addressing these uncertainties and proposing strategies for further improvement of current and novel NIs, NI-stabilized fertilizers could become a central tool for sustainable N management in agriculture.<br
Comparison of Complexity of Regular and Oblivious Decision Trees for Decision Tables with Many-valued Decisions from Closed Classes
No full textIn this paper, classes of decision tables with many-valued decisions (also known as multi-labeled decision tables) closed with respect to deletion of attributes (columns) and change of decisions are considered. For tables from these classes, the dependence of the minimum complexity of regular decision trees on the minimum complexity of oblivious decision trees, in which the order of queries on attribute values is predetermined, is studied. It is proved that the function describing this dependence is either bounded from above by a constant or grows as a logarithm, or grows almost linearly. This result is valid for the so-called bounded complexity measures, including the depth and weighted depth of decision trees. Motivation of this work constitutes in analyzing growth in the worst-case of the minimum complexity of a regular decision tree for a decision table from a closed class with the growth of the minimum complexity of an oblivious decision tree for this decision table. It is also shown that the minimum complexity of an oblivious tree for a decision table is equal to the minimum complexity of a reduct for this table. This finding can be useful for applications related to oblivious decision trees.ResearchreportedinthispublicationwassupportedbyKingAbdullahUniversityofScienceandTechnology (KAUST)
Temperature Drives Seagrass Recovery Across the Western North Atlantic
No full textClimate-driven shifts in herbivores, temperature, and nutrient runoff threaten coastal ecosystem resilience. However, ecological resilience, particularly for foundation species, remains poorly understood due to the scarcity of field experiments conducted across appropriate spatial and temporal scales that investigate multiple stressors. This study evaluates the resilience of a widespread tropical marine plant (turtlegrass) to disturbances across its geographic range and examines how environmental gradients in (a)biotic factors influence recovery. We assessed turtlegrass resilience by following recovery rates for a year after a simulated pulse disturbance (complete above- and belowground biomass removal). Contrary to studies in temperate areas, higher temperature generally enhanced seagrass recovery. While nutrients had minimal individual effects, they reduced aboveground recovery when combined with high levels of herbivore grazing (meso and megaherbivore). Belowground recovery was also affected by combined high levels of nutrients and grazing (megaherbivores only). Light availability had minimal effects. Our results suggest that the resilience of some tropical species, particularly in cooler subtropical waters, may initially benefit from warming. However, continuing shifts in nutrient supply and changes in grazing pressure may ultimately serve to compromise seagrass recovery.This work was supported by US National Science Foundation, OCE-1737116, OCE-1737144, OCE-1737247, OCE-2019022. Koninklijke Nederlandse Akademie van Wetenschappen, Ecology grant 2019. Nederlandse Organisatie voor Wetenschappelijk Onderzoek, 181.002
Sulfoxide-Functionalized Mesoporous C <sub>3</sub> N <sub>5</sub> as a Metal-Free and Visible-Light-Driven Efficient Photocatalyst for CO <sub>2</sub> Reduction
No full textPhotocatalytic CO2 reduction reaction provides a promising green strategy to mitigate atmospheric CO2 levels and simultaneously achieve valuable chemicals. C3N5 has emerged as an excellent metal-free photocatalyst for CO2RR due to its narrow bandgap with a lower conduction band edge, which provides high reduction potential to reduce thermodynamically stable CO2 molecules. Surface functionalization and enhanced porosity can yield superior catalytic performance in C3N5. The recent efforts to synthesize sulfoxide-functionalized mesoporous C3N5 is reported, which enhances CO2RR up to 4.4 times (94.8 ± 11.2 µmol·g−1 in 8 h) that of intact C3N5. According to Ultraviolet Photoelectron Spectroscopy analysis, while the functionalization of the C3N5 surface with sulfoxide groups leaves the valence band position unchanged, it significantly shifts the conduction band position toward a more negative potential, thereby enhancing material reduction capabilities. Additionally, steady-state photoluminescence and transient absorption and time-resolved photoluminescence spectra indicate a significant reduction in exciton recombination. Insights from theoretical calculations reveal that SO-functionalization enhances CO2 adsorption and CO desorption, thereby facilitating enhanced photocatalytic CO2 reduction.This research was supported by a grant from the National Research Foundation of Korea and funded by the Ministry of Science, ICT, and Future Planning. (Grant No. 2022K1A4A8A01080312). Financial support from the Slovenian Research and Innovation Agency (ARIS) through core funding P2-0152 (B.L.), project funding N1-0303, J2-4424 (M.H.) and J7-4638 (B.L.) and infrastructure funding I0-0039 (M.H.) is greatly appreciated. The computational resources were provided by the HPC RIVR consortium and EuroHPC JU through the HPC system Vega at the Institute of Information Science, Maribor, Slovenia. The authors also acknowledge the Sydney Analytical Center at the University of Sydney for providing them with access to their facilities and technical support for XPS, and UPS study
efunc: An Efficient Function Representation without Neural Networks
No full textFunction fitting/approximation plays a fundamental role in computer graphics and other engineering applications. While recent advances have explored neural networks to address this task, these methods often rely on architectures with many parameters, limiting their practical applicability. In contrast, we pursue high-quality function approximation using parameter-efficient representations that eliminate the dependency on neural networks entirely. We first propose a novel framework for continuous function modeling. Most existing works can be formulated using this framework. We then introduce a compact function representation, which is based on polynomials interpolated using radial basis functions, bypassing both neural networks and complex/hierarchical data structures. We also develop memory-efficient CUDA-optimized algorithms that reduce computational time and memory consumption to less than 10% compared to conventional automatic differentiation frameworks. Finally, we validate our representation and optimization pipeline through extensive experiments on 3D signed distance functions (SDFs). The proposed representation achieves comparable or superior performance to state-of-the-art techniques (e.g., octree/hash-grid techniques) with significantly fewer parameters
Distributed Memory Fast Fourier Transforms in the Exascale Era
No full textA summary of performance and software engineering concerns for the fast Fourier transform on distributed memory parallel computers is given. Index Terms—Fast Fourier Transform, Parallel software libraries, Computer performanc
SCRIPT: Predicting Single-Cell Long-Range Cis-Regulation Based on Pretrained Graph Attention Networks
No full textSingle-cell cis-regulatory relationships (CRRs) are essential for deciphering transcriptional regulation and understanding the pathogenic mechanisms of disease-associated non-coding variants. Existing computational methods struggle to accurately predict single-cell CRRs due to inadequately integrating causal biological principles and large-scale single-cell data. Here, SCRIPT (Single-cell Cis-regulatory Relationship Identifier based on Pre-Trained graph attention networks) is presented for inferring single-cell CRRs from transcriptomic and chromatin accessibility data. SCRIPT incorporates two key innovations: graph causal attention networks supported by empirical CRR evidence, and representation learning enhanced through pretraining on atlas-scale single-cell chromatin accessibility data. Validation using cell-type-specific chromatin contact and CRISPR perturbation data demonstrates that SCRIPT achieves a mean AUC of 0.89, significantly outperforming state-of-the-art methods (AUC: 0.7). Notably, SCRIPT obtains an over twofold improvement in predicting long-range CRRs (>100 Kb) compared to existing methods. By applying SCRIPT to Alzheimer's disease and schizophrenia, a framework is established for prioritizing disease-causing variants and elucidating their functional effects in a cell-type-specific manner. By uncovering molecular genetic mechanisms undetected by existing computational methods, SCRIPT provides a roadmap for advancing genetic diagnosis and target discovery.The authors gratefully acknowledge the following funding sources: National Natural Science Foundation of China (Grant No. 32370719,
32170667, 82394432, 92249302), Shanghai Municipal Science and Technology Major Project(2023SHZDZX02), the King Abdullah University
of Science and Technology (KAUST) Office of Research Administration (ORA) under Award No REI/1/5234-01-01, REI/1/5414-01-01, REI/1/5289-01-01, REI/1/5404-01-01, REI/1/5992-01-01, URF/1/4663-01-01, Center of Excellence for Smart Health (KCSH) under award number 5932, and Center of Excellence on Generative AI under award number 5940. The computations in this research were performed using the Computing for the Future at Fudan (CFFF) platform of Fudan University