Explore the vast range of titles available.

Inspired by the success of large language models (LLMs), we develop a long-context generative model for genomes. Our multiscale transformer model, megaDNA, is pre-trained on unannotated bacteriophage genomes with nucleotide-level tokenization. We demonstrate the foundational capabilities of our model including the prediction of essential genes, genetic variant effects, regulatory element activity and taxonomy of unannotated sequences. Furthermore, it generates de novo sequences up to 96 K base pairs, which contain potential regulatory elements and annotated proteins with phage-related functions. MegaDNA, a long-context genomic language model, generates DNA sequences up to 96 K base pairs with annotated proteins and potential regulatory elements. It predicts essential genes, genetic variant effects, and regulatory element activity.

The swift advancements in robotics have rendered navigation an essential task for mobile robots. While map-based navigation methods depend on global environmental maps for decision-making, their efficacy in unfamiliar or dynamic settings falls short. Current deep reinforcement learning navigation strategies can navigate successfully without pre-existing map data, yet they grapple with issues like inefficient training, slow convergence, and infrequent rewards. To tackle these challenges, this study introduces an improved two-delay depth deterministic policy gradient algorithm (LP-TD3) for local planning navigation. Initially, the integration of the long–short-term memory (LSTM) module with the Prioritized Experience Re-play (PER) mechanism into the existing TD3 framework was performed to optimize training and improve the efficiency of experience data utilization. Furthermore, the incorporation of an Intrinsic Curiosity Module (ICM) merges intrinsic with extrinsic rewards to tackle sparse reward problems and enhance exploratory behavior. Experimental evaluations using ROS and Gazebo simulators demonstrate that the proposed method outperforms the original on various performance metrics.

To achieve a stable size distribution over multiple generations, proliferating cells require a means of counteracting stochastic noise in the rate of growth, the time spent in various phases of the cell cycle, and the imprecision in the placement of the plane of cell division. In the most widely accepted model, cell size is thought to be regulated at the G1/S transition, such that cells smaller than a critical size pause at the end of G1 phase until they have accumulated mass to a predetermined size threshold, at which point the cells proceed through the rest of the cell cycle. However, a model, based solely on a specific size checkpoint at G1/S, cannot readily explain why cells with deficient G1/S control mechanisms are still able to maintain a very stable cell size distribution. Furthermore, such a model would not easily account for stochastic variation in cell size during the subsequent phases of the cell cycle, which cannot be anticipated at G1/S. To address such questions, we applied computationally enhanced quantitative phase microscopy (ceQPM) to populations of cultured human cell lines, which enables highly accurate measurement of cell dry mass of individual cells throughout the cell cycle. From these measurements, we have evaluated the factors that contribute to maintaining cell mass homeostasis at any point in the cell cycle. Our findings reveal that cell mass homeostasis is accurately maintained, despite disruptions to the normal G1/S machinery or perturbations in the rate of cell growth. Control of cell mass is generally not confined to regulation of the G1 length. Instead mass homeostasis is imposed throughout the cell cycle. In the cell lines examined, we find that the coefficient of variation (CV) in dry mass of cells in the population begins to decline well before the G1/S transition and continues to decline throughout S and G2 phases. Among the different cell types tested, the detailed response of cell growth rate to cell mass differs. However, in general, when it falls below that for exponential growth, the natural increase in the CV of cell mass is effectively constrained. We find that both mass-dependent cell cycle regulation and mass-dependent growth rate modulation contribute to reducing cell mass variation within the population. Through the interplay and coordination of these 2 processes, accurate cell mass homeostasis emerges. Such findings reveal previously unappreciated and very general principles of cell size control in proliferating cells. These same regulatory processes might also be operative in terminally differentiated cells. Further quantitative dynamical studies should lead to a better understanding of the underlying molecular mechanisms of cell size control.

Deep-learning-based object detection algorithms have significantly improved the performance of wheat spike detection. However, UAV images crowned with small-sized, highly dense, and overlapping spikes cause the accuracy to decrease for detection. This paper proposes an improved YOLOv5 (You Look Only Once)-based method to detect wheat spikes accurately in UAV images and solve spike error detection and miss detection caused by occlusion conditions. The proposed method introduces data cleaning and data augmentation to improve the generalization ability of the detection network. The network is rebuilt by adding a microscale detection layer, setting prior anchor boxes, and adapting the confidence loss function of the detection layer based on the IoU (Intersection over Union). These refinements improve the feature extraction for small-sized wheat spikes and lead to better detection accuracy. With the confidence weights, the detection boxes in multiresolution images are fused to increase the accuracy under occlusion conditions. The result shows that the proposed method is better than the existing object detection algorithms, such as Faster RCNN, Single Shot MultiBox Detector (SSD), RetinaNet, and standard YOLOv5. The average accuracy (AP) of wheat spike detection in UAV images is 94.1%, which is 10.8% higher than the standard YOLOv5. Thus, the proposed method is a practical way to handle the spike detection in complex field scenarios and provide technical references for field-level wheat phenotype monitoring.

Constructing powerful anode/cathode interphases by modulate ion solvation structure is the principle of electrolyte design. However, the methodological and theoretical design principles of electrolyte/solvation structure and their effect on electrochemical performance are still vague. Here, we propose a cationic weakly coordinating-intervention strategy for modulating the Na + solvation sheathes and constructing robust anode/cathode interphases in sodium-metal batteries. Unlike the local highly concentrated electrolytes, 1,2-difluorobenzene can weakly coordinate with Na + thus transforming the solvation structure into Na + -anion-incorporated structures and strengthening anode/cathode interphases formation by combining with salt decomposition. Furthermore, the correlations between the electrode interface properties and solvation structure are revealed, which can be tuned by the weakly coordination. Ultimately, the modulated electrolyte achieves 97.5% Coulombic efficiency for 600 cycles in Na‖Cu cells at 1 mA cm −2 and a beneficial lifetime (2500 h) in Na‖Na cells. Meanwhile, Na‖PB cells have achieved long-term operation at 4.8 V, along with operation at wide temperatures. Here, authors propose a weakly coordinating-intervention strategy to modulate the Na+ solvation sheath and construct a robust interphase in sodium-metal batteries. Correlation between electrode interface properties and solvation structure is also stated.

Ångstrom and attosecond are the fundamental spatiotemporal scales for electron dynamics in various materials. Although attosecond pulses with wavelengths comparable to the atomic scales are expected to be a key tool in advancing attosecond science, producing high-power hard X-ray attosecond pulses at ångstrom wavelengths remains a formidable challenge. Here, we report the generation of terawatt-scale attosecond hard X-ray pulses using a free-electron laser in a special operation mode. We achieved 9 keV single-spike X-ray pulses with a mean pulse energy of around 180 μJ, exceeding previous reports by more than an order of magnitude, and an estimated average pulse duration of 200 as at full-width at half-maximum. Exploiting the unique capability of the European XFEL, which can deliver ten pulse trains per second with each containing hundreds of pulses at megahertz repetition rates, this study demonstrates the generation of attosecond X-ray pulses at a 2.25 MHz repetition rate. These intense high-repetition-rate attosecond X-ray pulses present transformative prospects for structural and electronic damage-free X-ray measurements and attosecond time-resolved X-ray methodologies, heralding a new era in ultrafast X-ray science. Using the European XFEL free-electron laser, researchers demonstrate terawatt-scale, attosecond hard X-ray pulses. Ten pulse trains per second, each containing hundreds of pulses at megahertz repetition rates, are achieved. Such short and intense pulses at high repetition rate enable unprecedented damage-free X-ray measurements with attosecond temporal resolution.

Variational auto-encoders (VAE) have been widely used in process modeling due to the ability of deep feature extraction and noise robustness. However, the construction of a supervised VAE model still faces huge challenges. The data generated by the existing supervised VAE models are unstable and uncontrollable due to random resampling in the latent subspace, meaning the performance of prediction is greatly weakened. In this paper, a new multi-layer conditional variational auto-encoder (M-CVAE) is constructed by injecting label information into the latent subspace to control the output data generated towards the direction of the actual value. Furthermore, the label information is also used as the input with process variables in order to strengthen the correlation between input and output. Finally, a neural network layer is embedded in the encoder of the model to achieve online quality prediction. The superiority and effectiveness of the proposed method are demonstrated by two real industrial process cases that are compared with other methods.

In supercapacitors based on ionic liquid electrolytes, small amounts of absorbed water could potentially reduce the electrochemical window of electrolytes and cause performance degradation. The same would take place if ionic liquids are used as solvents for electrocatalysis involving the dissolved molecular species. In this work, we carry out molecular dynamics simulations, with gold and carbon electrodes in typical ionic liquids, hydrophobic and hydrophilic, to study electrosorption of water. We investigate the effects of hydrophobicity/hydrophilicity of ionic liquids and electrodes on interfacial distribution of ions and electrosorbed water. Results reveal that using hydrophilic ionic liquids would help to keep water molecules away from the negatively charged electrodes, even at large electrode polarizations. This conclusion is supported by electrochemical cyclic voltammetry measurements on gold and carbon electrodes in contact with humid ionic liquids. Thereby, our findings suggest potential mechanisms for protection of electrodes from water electrosorption. Ionic liquid electrolytes can impart increased operational voltage and energy density in supercapacitors, but water may diminish performance. Here the authors show that the hydrophilicity/hydrophobicity of ionic liquids can influence electrosorption of water and ultimately the supercapacitor performance.

Water-in-salt electrolytes are an appealing option for future electrochemical energy storage devices due to their safety and low toxicity. However, the physicochemical interactions occurring at the interface between the electrode and the water-in-salt electrolyte are not yet fully understood. Here, via in situ Raman spectroscopy and molecular dynamics simulations, we investigate the electrical double-layer structure occurring at the interface between a water-in-salt electrolyte and an Au(111) electrode. We demonstrate that most interfacial water molecules are bound with lithium ions and have zero, one, or two hydrogen bonds to feature three hydroxyl stretching bands. Moreover, the accumulation of lithium ions on the electrode surface at large negative polarizations reduces the interfacial field to induce an unusual “hydrogen-up” structure of interfacial water and blue shift of the hydroxyl stretching frequencies. These physicochemical behaviours are quantitatively different from aqueous electrolyte solutions with lower concentrations. This atomistic understanding of the double-layer structure provides key insights for designing future aqueous electrolytes for electrochemical energy storage devices. Water-in-salt electrolytes can be useful for future electrochemical energy storage systems. Here, the authors investigate the potential-dependent double-layer structures at the interface between a gold electrode and a highly concentrated aqueous electrolyte solution via in situ Raman measurements.

As the core component of solid-state batteries, neither current inorganic solid-state electrolytes nor solid polymer electrolytes can simultaneously possess satisfactory ionic conductivity, electrode compatibility and processability. By incorporating efficient Li + diffusion channels found in inorganic solid-state electrolytes and polar functional groups present in solid polymer electrolytes, it is conceivable to design inorganic-organic hybrid solid-state electrolytes to achieve true fusion and synergy in performance. Herein, we demonstrate that traditional metal coordination compounds can serve as exceptional Li + ion conductors at room temperature through rational structural design. Specifically, we synthesize copper maleate hydrate nanoflakes via bottom-up self-assembly featuring highly-ordered 1D channels that are interconnected by Cu 2+ /Cu + nodes and maleic acid ligands, alongside rich COO − groups and structural water within the channels. Benefiting from the combination of ion-hopping and coupling-dissociation mechanisms, Li + ions can preferably transport through these channels rapidly. Thus, the Li + -implanted copper maleate hydrate solid-state electrolytes shows remarkable ionic conductivity (1.17 × 10 −4  S cm −1 at room temperature), high Li + transference number (0.77), and a 4.7 V-wide operating window. More impressively, Li + -implanted copper maleate hydrate solid-state electrolytes are demonstrated to have exceptional compatibility with both cathode and Li anode, enabling long-term stability of more than 800 cycles. This work brings new insight on exploring superior room-temperature ionic conductors based on metal coordination compounds. The design of inorganic-organic hybrid solid-state electrolytes is expected to merge the merits of both inorganic and organic material. Here, the authors craft Li-ion-implanted copper maleate hydrate nanoflakes via a bottom-up self-assembly approach to reveal superior room-temperature Li-ion conductivity.