Search

Many disease-associated genomic variants disrupt gene function through abnormal splicing. With the advancement of genomic medicine, identifying disease-associated splicing associated variants has become more important than ever. Most bioinformatics approaches to detect splicing associated variants require both genome and transcriptomic data. However, there are not many datasets where both of them are available. In this study, we develop a methodology to detect genomic variants that cause splicing changes (more specifically, intron retention), using transcriptome sequencing data alone. After evaluating its sensitivity and precision, we apply it to 230,988 transcriptome sequencing data from the publicly available repository and identified 27,049 intron retention associated variants (IRAVs). In addition, by exploring positional relationships with variants registered in existing disease databases, we extract 3,000 putative disease-associated IRAVs, which range from cancer drivers to variants linked with autosomal recessive disorders. The in-silico screening framework demonstrates the possibility of near-automatically acquiring medical knowledge, making the most of massively accumulated publicly available sequencing data. Collections of IRAVs identified in this study are available through IRAVDB (https://iravdb.io/). This paper proposed a novel in-silico framework for automatically screening disease-related variants and applied it to over 200,000 transcriptomes, providing an example to acquire medically relevant knowledge from publicly available sequence data. [ABSTRACT FROM AUTHOR]

Low modulus materials that can shape-morph into different three-dimensional (3D) configurations in response to external stimuli have wide-ranging applications in flexible/stretchable electronics, surgical instruments, soft machines and soft robotics. This paper reports a shape-programmable system that exploits liquid metal microfluidic networks embedded in an elastomer matrix, with electromagnetic forms of actuation, to achieve a unique set of properties. Specifically, this materials structure is capable of fast, continuous morphing into a diverse set of continuous, complex 3D surfaces starting from a two-dimensional (2D) planar configuration, with fully reversible operation. Computational, multi-physics modeling methods and advanced 3D imaging techniques enable rapid, real-time transformations between target shapes. The liquid-solid phase transition of the liquid metal allows for shape fixation and reprogramming on demand. An unusual vibration insensitive, dynamic 3D display screen serves as an application example of this type of morphable surface. Low modulus materials that can change shape in response to external stimuli are promising for a wide range of applications. The authors here introduce a shape-reprogrammable construct, based on liquid metal microfluidic networks and electromagnetic actuation, that supports a unique collection of capabilities. [ABSTRACT FROM AUTHOR]

The activity of deep-focus earthquakes, which increases with depth from ~400 km to a peak at ~600 km, is enigmatic, because conventional brittle failure is unlikely to occur at elevated pressures. It becomes increasingly clear that pressure-induced phase transitions of olivine are responsible for the occurrence of the earthquakes, based on deformation experiments under pressure. However, many such experiments were made using analogue materials and those on mantle olivine are required to verify the hypotheses developed by these studies. Here we report the results of deformation experiments on (Mg,Fe)2SiO4 olivine at 11−17 GPa and 860−1350 K, equivalent to the conditions of the slabs subducted into the mantle transition zone. We find that throughgoing faulting occurs only at very limited temperatures of 1100−1160 K, accompanied by intense acoustic emissions at the onset of rupture. Fault sliding aided by shear heating occurs along a weak layer, which is formed via linking-up of lenticular packets filled with nanocrystalline olivine and wadsleyite. Our study suggests that transformational faulting occurs on the isothermal surface of the metastable olivine wedge in slabs, leading to deep-focus earthquakes in limited regions and depth range. This paper shows that formation of thin weak layers filled with nanocrystalline olivine/wadsleyite, upon the pressure-induced phase transition of olivine, is the major cause of deep-focus earthquakes on the metastable olivine wedge in deep slabs. [ABSTRACT FROM AUTHOR]

The problem of Lip-reading has become an important research challenge in recent years. The goal is to recognise speech from lip movements. Most of the Lip-reading technologies developed so far are camera-based, which require video recording of the target. However, these technologies have well-known limitations of occlusion and ambient lighting with serious privacy concerns. Furthermore, vision-based technologies are not useful for multi-modal hearing aids in the coronavirus (COVID-19) environment, where face masks have become a norm. This paper aims to solve the fundamental limitations of camera-based systems by proposing a radio frequency (RF) based Lip-reading framework, having an ability to read lips under face masks. The framework employs Wi-Fi and radar technologies as enablers of RF sensing based Lip-reading. A dataset comprising of vowels A, E, I, O, U and empty (static/closed lips) is collected using both technologies, with a face mask. The collected data is used to train machine learning (ML) and deep learning (DL) models. A high classification accuracy of 95% is achieved on the Wi-Fi data utilising neural network (NN) models. Moreover, similar accuracy is achieved by VGG16 deep learning model on the collected radar-based dataset. This study aims to solve the fundamental limitations of camera-based systems by proposing a contactless and privacy-preserving radio frequency-based lip-reading framework, which can recognise lip movements under face masks. [ABSTRACT FROM AUTHOR]

Ultraviolet colouration is thought to be an important form of signalling in many bird species, yet broad insights regarding the prevalence of ultraviolet plumage colouration and the factors promoting its evolution are currently lacking. In this paper, we develop a image segmentation pipeline based on deep learning that considerably outperforms classical (i.e. non deep learning) segmentation methods, and use this to extract accurate information on whole-body plumage colouration from photographs of >24,000 museum specimens covering >4500 species of passerine birds. Our results demonstrate that ultraviolet reflectance, particularly as a component of other colours, is widespread across the passerine radiation but is strongly phylogenetically conserved. We also find clear evidence in support of the role of light environment in promoting the evolution of ultraviolet plumage colouration, and a weak trend towards higher ultraviolet plumage reflectance among bird species with ultraviolet rather than violet-sensitive visual systems. Overall, our study provides important broad-scale insight into an enigmatic component of avian colouration, as well as demonstrating that deep learning has considerable promise for allowing new data to be brought to bear on long-standing questions in ecology and evolution. Here, the authors develop software that uses photographs of birds to extract information on plumage UV reflectance. They use these data to show that UV reflectance is phylogenetically conserved and associated with the light environment. [ABSTRACT FROM AUTHOR]

Tool based manufacturing processes like injection moulding allow fast and high-quality mass-market production, but for optical polymer components the production of the necessary tools is time-consuming and expensive. In this paper a process to fabricate metal-inserts for tool based manufacturing with smooth surfaces via a casting and replication process from fused silica templates is presented. Bronze, brass and cobalt-chromium could be successfully replicated from shaped fused silica replications achieving a surface roughnesses of Rq 8 nm and microstructures in the range of 5 µm. Injection moulding was successfully performed, using a commercially available injection moulding system, with thousands of replicas generated from the same tool. In addition, three-dimensional bodies in metal could be realised with 3D-Printing of fused silica casting moulds. This work thus represents an approach to high-quality moulding tools via a scalable facile and cost-effective route surpassing the currently employed cost-, labour- and equipment-intensive machining techniques. Production of tools for polymer replication in the field of optical applications is still time-consuming and cost-intensive. Here the authors develop an efficient metal casting process, and demonstrate manufacturing of structures of complex shapes with a surface roughness of few nanometres. [ABSTRACT FROM AUTHOR]

While inequalities in science are common, most efforts to understand them treat scientists as isolated individuals, ignoring the network effects of collaboration. Here, we develop models that untangle the network effects of productivity defined as paper counts, and prominence referring to high-impact publications, of individual scientists from their collaboration networks. We find that gendered differences in the productivity and prominence of mid-career researchers can be largely explained by differences in their coauthorship networks. Hence, collaboration networks act as a form of social capital, and we find evidence of their transferability from senior to junior collaborators, with benefits that decay as researchers age. Collaboration network effects can also explain a large proportion of the productivity and prominence advantages held by researchers at prestigious institutions. These results highlight a substantial role of social networks in driving inequalities in science, and suggest that collaboration networks represent an important form of unequally distributed social capital that shapes who makes what scientific discoveries. While inequalities in science are common, most efforts to understand them treat scientists as isolated individuals, ignoring the network effects of collaboration. Here, the authors develop models that untangle the network effects of productivity and prominence of individual scientists from their collaboration networks. [ABSTRACT FROM AUTHOR]

In order to realize the full potential of wireless edge artificial intelligence (AI), very large and diverse datasets will often be required for energy-demanding model training on resource-constrained edge devices. This paper proposes a lead federated neuromorphic learning (LFNL) technique, which is a decentralized energy-efficient brain-inspired computing method based on spiking neural networks. The proposed technique will enable edge devices to exploit brain-like biophysiological structure to collaboratively train a global model while helping preserve privacy. Experimental results show that, under the situation of uneven dataset distribution among edge devices, LFNL achieves a comparable recognition accuracy to existing edge AI techniques, while substantially reducing data traffic by >3.5× and computational latency by >2.0×. Furthermore, LFNL significantly reduces energy consumption by >4.5× compared to standard federated learning with a slight accuracy loss up to 1.5%. Therefore, the proposed LFNL can facilitate the development of brain-inspired computing and edge AI. Designing energy-efficient computing solution for the implementation of AI algorithms in edge devices remains a challenge. Yang et al. proposes a decentralized brain-inspired computing method enabling multiple edge devices to collaboratively train a global model without a fixed central coordinator. [ABSTRACT FROM AUTHOR]

In the context of visual attention, it has been classically assumed that missing the response to a target or erroneously selecting a distractor occurs as a consequence of the (miss)allocation of attention in space. In the present paper, we challenge this view and provide evidence that, in addition to encoding spatial attention, prefrontal neurons also encode a distractibility-to-impulsivity state. Using supervised dimensionality reduction techniques in prefrontal neuronal recordings in monkeys, we identify two partially overlapping neuronal subpopulations associated either with the focus of attention or overt behaviour. The degree of overlap accounts for the behavioral gain associated with the good allocation of attention. We further describe the neural variability accounting for distractibility-to-impulsivity behaviour by a two dimensional state associated with optimality in task and responsiveness. Overall, we thus show that behavioral performance arises from the integration of task-specific neuronal processes and pre-existing neuronal states describing task-independent behavioral states. Failing to detect relevant information has been assumed to be a consequence of misallocation of attention. Here, the authors present findings showing that optimal behavioral performance results from the absence of interference between internal neural states and attention control. [ABSTRACT FROM AUTHOR]

Near-zero-index (NZI) media have been theoretically identified as media where electromagnetic radiations behave like ideal electromagnetic fluids. Within NZI media, the electromagnetic power flow obeys equations similar to those of motion for the velocity field in an ideal fluid, so that optical turbulence is intrinsically inhibited. Here, we experimentally observe the electromagnetic power flow distribution of such an ideal electromagnetic fluid propagating within a cutoff waveguide by a semi-analytical reconstruction technique. This technique provides direct proof of the inhibition of electromagnetic vorticity at the NZI frequency, even in the presence of complex obstacles and topological changes in the waveguide. Phase uniformity and spatially-static field distributions, essential characteristics of NZI materials, are also observed. Measurement of the same structure outside the NZI frequency range reveals existence of vortices in the power flow, as expected for conventional optical systems. Therefore, our results provide an important step forward in the development of ideal electromagnetic fluids, and introduce a tool to explore the subwavelength behavior of NZI media including fully vectorial and phase information. This paper reports an experimental observation of irrotational, inviscid, and incompressible electromagnetic power flow within an epsilon-near-zero medium, exhibiting an analogy to an ideal fluid. [ABSTRACT FROM AUTHOR]

The orderly arrangement of nanomaterials' tiny units at the nanometer-scale accounts for a substantial part of their remarkable properties. Maintaining this orderness and meanwhile endowing the nanomaterials with highly precise and free-designed 3D micro architectures will open an exciting prospect for various novel applications. In this paper, we developed a sacrificial-scaffold-mediated two-photon lithography (TPL) strategy that enables the fabrication of complex 3D colloidal crystal microstructures with orderly-arranged nanoparticles inside. We show that, with the help of a degradable hydrogel scaffold, the disturbance effect of the femtosecond laser to the nanoparticle self-assembling could be overcome. Therefore, hydrogel-state and solid-state colloidal crystal microstructures with diverse compositions, free-designed geometries and variable structural colors could be easily fabricated. This enables the possibility to create novel colloidal crystal microsensing systems that have not been achieved before. Colloidal crystals are widely applied in the fabrication of optoelectronic devices, but realizing freedom of design, such as in 3D printing, in colloidal crystal fabrication remains challenging. Here, the authors demonstrate a sacrificial-scaffold-mediated two-photon lithography strategy that enables the fabrication of complex 3D colloidal crystal microstructures with orderly arranged nanoparticles in the bulk. [ABSTRACT FROM AUTHOR]

Protective layers are essential for Si-based photocathodes to achieve long-term stability. The conventionally used inorganic protective layers, such as TiO2, need to be free of pinholes to isolate Si from corrosive solution, which demands extremely high-quality deposition techniques. On the other hand, organic hydrophobic protective layers suffer from the trade-off between current density and stability. This paper describes the design and fabrication of a discontinuous hybrid organic protective layer with controllable surface wettability. The underlying hydrophobic layer induces the formation of thin gas layers at the discontinuous pores to isolate the electrolyte from Si substrate, while allowing Pt co-catalyst to contact the electrolyte for water splitting. Meanwhile, the surface of this organic layer is modified with hydrophilic hydroxyl groups to facilitate bubble detachment. The optimized photocathode achieves a stable photocurrent of 35 mA/cm2 for over 110 h with no trend of decay. Preparation of inorganic protective layers for photoelectrodes requires high-quality deposition techniques. Here, the authors report a spin-coated organic protective layer that enables Si photocathodes to realize stable solar water splitting. [ABSTRACT FROM AUTHOR]

Ferroelectrics are considered excellent photocatalytic candidates for solar fuel production because of the unidirectional charge separation and above-gap photovoltage. Nevertheless, the performance of ferroelectric photocatalysts is often moderate. A few studies showed that these types of photocatalysts could achieve overall water splitting. This paper proposes an approach to fabricating interfacial charge-collecting nanostructures on positive and negative domains of ferroelectric, enabling water splitting in ferroelectric photocatalysts. The present study observes efficient accumulations of photogenerated electrons and holes within their thermalization length (~50 nm) around Au nanoparticles located in the positive and negative domains of a BaTiO3 single crystal. Photocatalytic overall water splitting is observed on a ferroelectric BaTiO3 single crystal after assembling oxidation and reduction cocatalysts on the positively and negatively charged Au nanoparticles, respectively. The fabrication of bipolar charge-collecting structures on ferroelectrics to achieve overall water splitting offers a way to utilize the energetic photogenerated charges in solar energy conversion. While ferroelectric materials are promising candidates for solar water splitting, most examples show poor activities. Here, authors prepare charge-collecting nanostructures on the positive and negative domains of BaTiO3 and demonstrate photocatalytic overall water splitting. [ABSTRACT FROM AUTHOR]

Here we show how major rivers can efficiently connect to the deep-sea, by analysing the longest runout sediment flows (of any type) yet measured in action on Earth. These seafloor turbidity currents originated from the Congo River-mouth, with one flow travelling >1,130 km whilst accelerating from 5.2 to 8.0 m/s. In one year, these turbidity currents eroded 1,338-2,675 [>535-1,070] Mt of sediment from one submarine canyon, equivalent to 19–37 [>7–15] % of annual suspended sediment flux from present-day rivers. It was known earthquakes trigger canyon-flushing flows. We show river-floods also generate canyon-flushing flows, primed by rapid sediment-accumulation at the river-mouth, and sometimes triggered by spring tides weeks to months post-flood. It is demonstrated that strongly erosional turbidity currents self-accelerate, thereby travelling much further, validating a long-proposed theory. These observations explain highly-efficient organic carbon transfer, and have important implications for hazards to seabed cables, or deep-sea impacts of terrestrial climate change. This paper analyses the longest sediment flows measured in action on Earth. These seabed flows were caused by floods and spring tides, and flushed prodigious sediment and carbon volumes into the deep sea, as they accelerated for a thousand kilometres. [ABSTRACT FROM AUTHOR]

Monolayer 2D semiconductors provide an attractive option for valleytronics due to valley-addressability. But the short valley-polarization lifetimes for excitons have hindered potential valleytronic applications. In this paper, we demonstrate a strategy for prolonging the valley-polarization lifetime by converting excitons to trions through efficient gate control and exploiting the much longer valley-polarization lifetimes for trions than for excitons. At charge neutrality, the valley lifetime of monolayer MoTe2 increases by a factor of 1000 to the order of nanoseconds from excitons to trions. The exciton-to-trion conversion changes the dominant depolarization mechanism from the fast electron-hole exchange for excitons to the slow spin-flip process for trions. Moreover, the degree of valley polarization increases to 38% for excitons and 33% for trions through electrical manipulation. Our results reveal the depolarization dynamics and the interplay of various depolarization channels for excitons and trions, providing an effective strategy for prolonging the valley polarization. Here, the authors devise a strategy for prolonging the valley polarization lifetime in monolayer MoTe2 by converting excitons to trions through gate control, and by taking advantage of the longer valley polarization lifetime of trions. [ABSTRACT FROM AUTHOR]

Little is known about the structural basis of prion strains. Here we provide a high (3.0 Å) resolution cryo-electron microscopy-based structure of infectious brain-derived fibrils of the mouse anchorless RML scrapie strain which, like the recently determined hamster 263K strain, has a parallel in-register β-sheet-based core. Several structural motifs are shared between these ex vivo prion strains, including an amino-proximal steric zipper and three β-arches. However, detailed comparisons reveal variations in these shared structural topologies and other features. Unlike 263K and wildtype RML prions, the anchorless RML prions lack glycophosphatidylinositol anchors and are severely deficient in N-linked glycans. Nonetheless, the similarity of our anchorless RML structure to one reported for wildtype RML prion fibrils in an accompanying paper indicates that these post-translational modifications do not substantially alter the amyloid core conformation. This work demonstrates both common and divergent structural features of prion strains at the near-atomic level. The authors report the near-atomic structure of the ex vivo aRML prion fibril. The comparison between this structure and the previously solved 263K prion fibril shows that both common and divergent features characterize these prion strains and their respective conformational replication templates. [ABSTRACT FROM AUTHOR]

This paper is a call to action. By publishing concurrently across journals like an emergency bulletin, we are not merely making a plea for awareness about climate change. Instead, we are demanding immediate, tangible steps that harness the power of microbiology and the expertise of researchers and policymakers to safeguard the planet for future generations.

In the initial version of our paper, we included Clean Water Act data for all 50 states, DC, and Puerto Rico. Similarly, the percent of Clean Water Act permittees in Significant Noncompliance drops from 6.01 to 3.37% and the average percent of permittees in Significant Noncompliance drops from 9.00 to 6.23%. To ensure that we did not misrepresent the level of Clean Water Act Significant Noncompliance in these states, we have corrected our article by removing these states from the Clean Water Act portion of our analysis. [Extracted from the article]

An amendment to this paper has been published and can be accessed via a link at the top of the paper. [ABSTRACT FROM AUTHOR]

An amendment to this paper has been published and can be accessed via a link at the top of the paper. [ABSTRACT FROM AUTHOR]

An amendment to this paper has been published and can be accessed via a link at the top of the paper. [ABSTRACT FROM AUTHOR]

An amendment to this paper has been published and can be accessed via a link at the top of the paper. [ABSTRACT FROM AUTHOR]

An amendment to this paper has been published and can be accessed via a link at the top of the paper. [ABSTRACT FROM AUTHOR]

Abstract Cash transfer policies have been widely discussed as mechanisms to curb intergenerational transmission of socioeconomic disadvantage. In this paper, we take advantage of a large casino-funded family transfer program introduced in a Southeastern American Indian Tribe to generate difference-in-difference estimates of the link between children’s cash transfer exposure and third grade math and reading test scores of their offspring. Here we show greater math (0.25 standard deviation [SD], p =.0148, 95% Confidence Interval [CI]: 0.05, 0.45) and reading (0.28 SD, p = .0066, 95% CI: 0.08, 0.49) scores among American Indian students whose mother was exposed ten years longer than other American Indian students to the cash transfer during her childhood (or relative to the non-American Indian student referent group). Exploratory analyses find that a mother’s decision to pursue higher education and delay fertility appears to explain some, but not all, of the relation between cash transfers and children’s test scores. In this rural population, large cash transfers have the potential to reduce intergenerational cycles of poverty-related educational outcomes.

Abstract Low-cost, efficient catalyst high-throughput screening is crucial for future renewable energy technology. Interpretable machine learning is a powerful method for accelerating catalyst design by extracting physical meaning but faces huge challenges. This paper describes an interpretable descriptor model to unify activity and selectivity prediction for multiple electrocatalytic reactions (i.e., O2/CO2/N2 reduction and O2 evolution reactions), utilizing only easily accessible intrinsic properties. This descriptor, named ARSC, successfully decouples the atomic property (A), reactant (R), synergistic (S), and coordination effects (C) on the d-band shape of dual-atom sites, which is built upon our developed physically meaningful feature engineering and feature selection/sparsification (PFESS) method. Driven by this descriptor, we can rapidly locate optimal catalysts for various products instead of over 50,000 density functional theory calculations. The model’s universality has been validated by abundant reported works and subsequent experiments, where Co-Co/Ir-Qv3 are identified as optimal bifunctional oxygen reduction and evolution electrocatalysts. This work opens the avenue for intelligent catalyst design in high-dimensional systems linked with physical insights.

Abstract DNA data storage is a potential alternative to magnetic tape for archival storage purposes, promising substantial gains in information density. Critical to the success of DNA as a storage media is an understanding of the role of environmental factors on the longevity of the stored information. In this paper, we evaluate the effect of exposure to ionizing particle radiation, a cause of data loss in traditional magnetic media, on the longevity of data in DNA data storage pools. We develop a mass action kinetics model to estimate the rate of damage accumulation in DNA strands due to neutron interactions with both nucleotides and residual water molecules, then utilize the model to evaluate the effect several design parameters of a typical DNA data storage scheme have on expected data longevity. Finally, we experimentally validate our model by exposing dried DNA samples to different levels of neutron irradiation and analyzing the resulting error profile. Our results show that particle radiation is not a significant contributor to data loss in DNA data storage pools under typical storage conditions.

Abstract Spatiotemporal datasets, which consist of spatially-referenced time series, are ubiquitous in diverse applications, such as air pollution monitoring, disease tracking, and cloud-demand forecasting. As the scale of modern datasets increases, there is a growing need for statistical methods that are flexible enough to capture complex spatiotemporal dynamics and scalable enough to handle many observations. This article introduces the Bayesian Neural Field (BayesNF), a domain-general statistical model that infers rich spatiotemporal probability distributions for data-analysis tasks including forecasting, interpolation, and variography. BayesNF integrates a deep neural network architecture for high-capacity function estimation with hierarchical Bayesian inference for robust predictive uncertainty quantification. Evaluations against prominent baselines show that BayesNF delivers improvements on prediction problems from climate and public health data containing tens to hundreds of thousands of measurements. Accompanying the paper is an open-source software package ( https://github.com/google/bayesnf ) that runs on GPU and TPU accelerators through the Jax machine learning platform.

Abstract Several gaps persist in haptic device development due to the multifaceted nature of the sense of touch. Existing gaps include challenges enhancing touch feedback fidelity, providing diverse haptic sensations, and ensuring wearability for delivering tactile stimuli to the fingertips. Here, we introduce the Bioinspired Adaptable Multiplanar Haptic system, offering mechanotactile/steady and vibrotactile pulse stimuli with adjustable intensity (up to 298.1 mN) and frequencies (up to 130 Hz). This system can deliver simultaneous stimuli across multiple fingertip areas. The paper includes a full characterisation of our system. As the device can play an important role in further understanding human touch, we performed human stimuli sensitivity and differentiation experiments to evaluate the capability of delivering mechano-vibrotactile, variable intensity, simultaneous, multiplanar and operator agnostic stimuli. Our system promises to accelerate the development of touch perception devices, providing painless, operator-independent data crucial for researching and diagnosing touch-related disorders.

Abstract Increasing the binding affinity of an antibody to its target antigen is a crucial task in antibody therapeutics development. This paper presents a pretrainable geometric graph neural network, GearBind, and explores its potential in in silico affinity maturation. Leveraging multi-relational graph construction, multi-level geometric message passing and contrastive pretraining on mass-scale, unlabeled protein structural data, GearBind outperforms previous state-of-the-art approaches on SKEMPI and an independent test set. A powerful ensemble model based on GearBind is then derived and used to successfully enhance the binding of two antibodies with distinct formats and target antigens. ELISA EC50 values of the designed antibody mutants are decreased by up to 17 fold, and K D values by up to 6.1 fold. These promising results underscore the utility of geometric deep learning and effective pretraining in macromolecule interaction modeling tasks.

Abstract Climate change is projected to have substantial economic, social, and environmental impacts worldwide. Currently, the leading solutions for hydrogen storage are in salt caverns, and depleted natural gas reservoirs. However, the required geological formations are limited to certain regions. To increase alternatives for hydrogen storage, this paper proposes storing hydrogen in pipes filled with gravel in lakes, hydropower, and pumped hydro storage reservoirs. Hydrogen is insoluble in water, non-toxic, and does not threaten aquatic life. Results show the levelized cost of hydrogen storage to be 0.17 USD kg−1 at 200 m depth, which is competitive with other large scale hydrogen storage options. Storing hydrogen in lakes, hydropower, and pumped hydro storage reservoirs increases the alternatives for storing hydrogen and might support the development of a hydrogen economy in the future. The global potential for hydrogen storage in reservoirs and lakes is 3 and 12 PWh, respectively. Hydrogen storage in lakes and reservoirs can support the development of a hydrogen economy in the future by providing abundant and cheap hydrogen storage.

Abstract Today’s wireless environments, soft robotics, and space applications demand delicate design of devices with tunable performances and simple fabrication processes. Here we show strain-based adjustability of RF/microwave performance by applying frequency-selective patterns of conductive Ti3C2T x MXene coatings on low-cost acetate substrates under ambient conditions. The tailored performances were achieved by applying frequency-selective patterns of thin Ti3C2Tx MXene coatings with high electrical conductivity as a replacement to metal on low-cost flexible acetate substrates under ambient conditions. Under quasi-axial stress, the Kirigami design enables displacements of individual resonant cells, changing the overall electromagnetic performance of a surface (i.e., array) within a simulated wireless channel. Two flexible Kirigami-inspired prototypes were implemented and tested within the S, C, and X (2-4 GHz, 4-8 GHz, and 8-12 GHz) microwave frequency bands. The resonant surface, having ~1/4 of the size of a standard A4 paper, was able to steer a beam of scattered waves from each resonator by ~25°. Under a strain of 22%, the resonant frequency of the wired co-planar resonator was shifted by 400 MHz, while the reflection coefficient changed by 158%. Deforming the geometry impacted the spectral response of the components across three arbitrary frequencies in the 4-10 GHz frequency range. With this proof of concept, we anticipate implementing thin films of MXenes on technologically relevant substrates, achieving multi-functionality through cost-effective and straightforward manufacturing.

Abstract The fracture behaviour of artificial metamaterials often leads to catastrophic failures with limited resistance to crack propagation. In contrast, natural materials such as bones and ceramics possess microstructures that give rise to spatially controllable crack path and toughened material resistance to crack advances. This study presents an approach that is inspired by nature’s strengthening mechanisms to develop a systematic design method enabling damage-programmable metamaterials with engineerable microfibers in the cells that can spatially program the micro-scale crack behaviour. Machine learning is applied to provide an effective design engine that accelerate the generation of damage-programmable cells that offer advanced toughening functionality such as crack bowing, crack deflection, and shielding seen in natural materials; and are optimised for a given programming of crack path. This paper shows that such toughening features effectively enable crack-resisting mechanisms on the basis of the crack tip interactions, crack shielding, crack bridging and synergistic combinations of these mechanisms, increasing up to 1,235% absorbed fracture energy in comparison to conventional metamaterials. The proposed approach can have broad implications in the design of damage-tolerant materials, and lightweight engineering systems where significant fracture resistances or highly programmable damages for high performances are sought after.

Abstract Quantum reservoir computing (QRC) has been proposed as a paradigm for performing machine learning with quantum processors where the training takes place in the classical domain, avoiding the issue of barren plateaus in parameterized-circuit quantum neural networks. It is natural to consider using a quantum processor based on microwave superconducting circuits to classify microwave signals that are analog—continuous in time. However, while there have been theoretical proposals of analog QRC, to date QRC has been implemented using the circuit model—imposing a discretization of the incoming signal in time. In this paper we show how a quantum superconducting circuit comprising an oscillator coupled to a qubit can be used as an analog quantum reservoir for a variety of classification tasks, achieving high accuracy on all of them. Our work demonstrates processing of ultra-low-power microwave signals within our superconducting circuit, a step towards achieving a quantum sensing-computational advantage on impinging microwave signals.

Abstract Microstructures of additively manufactured metal parts are crucial since they determine the mechanical properties. The evolution of the microstructures during layer-wise printing is complex due to continuous re-melting and reheating effects. The current approach to studying this phenomenon relies on time-consuming numerical models such as finite element analysis due to the lack of effective sub-surface temperature measurement techniques. Attributed to the miniature footprint, chirped-fiber Bragg grating, a unique type of fiber optical sensor, has great potential to achieve this goal. However, using the traditional demodulation methods, its spatial resolution is limited to the millimeter level. In addition, embedding it during laser additive manufacturing is challenging since the sensor is fragile. This paper implements a machine learning-assisted approach to demodulate the optical signal to thermal distribution and significantly improve spatial resolution to 28.8 µm from the original millimeter level. A sensor embedding technique is also developed to minimize damage to the sensor and part while ensuring close contact. The case study demonstrates the excellent performance of the proposed sensor in measuring sharp thermal gradients and fast cooling rates during the laser powder bed fusion. The developed sensor has a promising potential to study the fundamental physics of metal additive manufacturing processes.

Abstract Micro-/nano-robots (MNRs) have impressive potential in minimally invasive targeted therapeutics through blood vessels, which has disruptive impact to improving human health. However, the clinical use of MNRs has yet to happen due to intrinsic limitations, such as overcoming blood flow. These bottlenecks have not been empirically solved. To tackle them, a full understanding of MNR behaviors is necessary as the first step. The common movement principle of MNRs is corkscrew motion with a helical structure. The existing dynamic model is only applicable to standard helical MNRs. In this paper, we propose a dynamic model for general MNRs without structure limitations. Comprehensive simulations and experiments were conducted, which shows the validity and accuracy of our model. Such a model can serve as a reliable basis for the design, optimization, and control of MNRs and as a powerful tool for gaining fluid dynamic insights, thus accelerating the development of the field.

Abstract Free-electron-lasers fill a critical gap in the space of THz-sources as they can reach high average and peak powers with spectral tunability. Using a waveguide in a THz FEL significantly increases the coupling between the relativistic electrons and electromagnetic field enabling large amounts of radiation to be generated in a single passage of electrons through the undulator. In addition to transversely confining the radiation, the dispersive properties of the waveguide critically affect the velocity and slippage of the radiation pulse which determine the central frequency and bandwidth of the generated radiation. In this paper, we characterize the spectral properties of a compact waveguide THz FEL including simultaneous lasing at two different frequencies and demonstrating tuning of the radiation wavelength in the high frequency branch by varying the beam energy and ensuring that the electrons injected into the undulator are prebunched on the scale of the resonant radiation wavelength.

Abstract Single-cell RNA sequencing (scRNA-seq) technologies have become essential tools for characterizing cellular landscapes within complex tissues. Large-scale single-cell transcriptomics holds great potential for identifying rare cell types critical to the pathogenesis of diseases and biological processes. Existing methods for identifying rare cell types often rely on one-time clustering using partial or global gene expression. However, these rare cell types may be overlooked during the clustering phase, posing challenges for their accurate identification. In this paper, we propose a Cluster decomposition-based Anomaly Detection method (scCAD), which iteratively decomposes clusters based on the most differential signals in each cluster to effectively separate rare cell types and achieve accurate identification. We benchmark scCAD on 25 real-world scRNA-seq datasets, demonstrating its superior performance compared to 10 state-of-the-art methods. In-depth case studies across diverse datasets, including mouse airway, brain, intestine, human pancreas, immunology data, and clear cell renal cell carcinoma, showcase scCAD’s efficiency in identifying rare cell types in complex biological scenarios. Furthermore, scCAD can correct the annotation of rare cell types and identify immune cell subtypes associated with disease, thereby offering valuable insights into disease progression.

Abstract Inspired by the natural shape-morphing abilities of biological organisms, we introduce a strategy for creating energy-efficient dynamic 3D metasurfaces through spatiotemporal jamming of interleaved assemblies. Our approach, diverging from traditional shape-morphing techniques reliant on continuous energy inputs, utilizes strategically jammed, paper-based interleaved assemblies. By rapidly altering their stiffness at various spatial points and temporal phases during the relaxation of the soft substrate through jamming, we enable the formation of refreshable, intricate 3D shapes with a desirable load-bearing capability. This process, which does not require ongoing energy consumption, ensures energy-efficient and lasting shape displays. Our theoretical model, linking buckling deformation to residual pre-strain, underpins the inverse design process for an array of interleaved assemblies, facilitating the creation of diverse 3D configurations. This metasurface holds notable potential for tactile displays, particularly for the visually impaired, heralding possibilities in visual impaired education, haptic feedback, and virtual/augmented reality applications.

Abstract Synthetic droplets mimicking bio-soft matter droplets formed via liquid-liquid phase separation (LLPS) in living cells have recently been employed in nanobiotechnology for artificial cells, molecular robotics, molecular computing, etc. Temporally controlling the dynamics of synthetic droplets is essential for developing such bio-inspired systems because living systems maintain their functions based on the temporally controlled dynamics of biomolecular reactions and assemblies. This paper reports the temporal control of DNA-based LLPS droplets (DNA droplets). We demonstrate the timing-controlled division of DNA droplets via time-delayed division triggers regulated by chemical reactions. Controlling the release order of multiple division triggers results in order control of the multistep droplet division, i.e., pathway-controlled division in a reaction landscape. Finally, we apply the timing-controlled division into a molecular computing element to compare microRNA concentrations. We believe that temporal control of DNA droplets will promote the design of dynamic artificial cells/molecular robots and sophisticated biomedical applications.

Abstract The discovery of ambient superconductivity would mark an epochal breakthrough long-awaited for over a century, potentially ushering in unprecedented scientific and technological advancements. The recent findings on high-temperature superconducting phases in various hydrides under high pressure have ignited optimism, suggesting that the realization of near-ambient superconductivity might be on the horizon. However, the preparation of hydride samples tends to promote the emergence of various metastable phases, marked by a low level of experimental reproducibility. Identifying these phases through theoretical and computational methods entails formidable challenges, often resulting in controversial outcomes. In this paper, we consider N-doped LuH3 as a prototypical complex hydride: By means of machine-learning-accelerated force-field molecular dynamics, we have identified the formation of H2 molecules stabilized at ambient pressure by nitrogen impurities. Importantly, we demonstrate that this molecular phase plays a pivotal role in the emergence of a dynamically stable, low-temperature, experimental-ambient-pressure superconductivity. The potential to stabilize hydrogen in molecular form through chemical doping opens up a novel avenue for investigating disordered phases in hydrides and their transport properties under near-ambient conditions.

Abstract This paper employs insights from earth science on the financial risk of project developments to present an economic theory of critical minerals. Our theory posits that back-ended critical mineral projects that have unaddressed technical and non-technical barriers, such as those involving lithium and cobalt, exhibit an additional risk for investors which we term the “back-ended risk premium”. We show that the back-ended risk premium increases the cost of capital and, therefore, has the potential to reduce investment in the sector. We posit that the back-ended risk premium may also reduce the gains in productivity expected from artificial intelligence (AI) technologies in the mining sector. Progress in AI may, however, lessen the back-ended risk premium itself by shortening the duration of mining projects and the required rate of investment by reducing the associated risk. We conclude that the best way to reduce the costs associated with energy transition is for governments to invest heavily in AI mining technologies and research.

Abstract Point-of-care serological and direct antigen testing offers actionable insights for diagnosing challenging illnesses, empowering distributed health systems. Here, we report a POC-compatible serologic test for Lyme disease (LD), leveraging synthetic peptides specific to LD antibodies and a paper-based platform for rapid, and cost-effective diagnosis. Antigenic epitopes conserved across Borrelia burgdorferi genospecies, targeted by IgG and IgM antibodies, are selected to develop a multiplexed panel for detection of LD antibodies from patient sera. Multiple peptide epitopes, when combined synergistically with a machine learning-based diagnostic model achieve high sensitivity without sacrificing specificity. Blinded validation with 15 LD-positive and 15 negative samples shows 95.5% sensitivity and 100% specificity. Blind testing with the CDC’s LD repository samples confirms the test accuracy, matching lab-based two-tier results, correctly differentiating between LD and look-alike diseases. This LD diagnostic test could potentially replace the cumbersome two-tier testing, improving diagnosis and enabling earlier treatment while facilitating immune monitoring and surveillance.

Abstract As robots are increasingly participating in our daily lives, the quests to mimic human abilities have driven the advancements of robotic multimodal senses. However, current perceptual technologies still unsatisfied robotic needs for home tasks/environments, particularly facing great challenges in multisensory integration and fusion, rapid response capability, and highly sensitive perception. Here, we report a flexible tactile sensor utilizing thin-film thermistors to implement multimodal perceptions of pressure, temperature, matter thermal property, texture, and slippage. Notably, the tactile sensor is endowed with an ultrasensitive (0.05 mm/s) and ultrafast (4 ms) slip sensing that is indispensable for dexterous and reliable grasping control to avoid crushing fragile objects or dropping slippery objects. We further propose and develop a robotic tactile-visual fusion architecture that seamlessly encompasses multimodal sensations from the bottom level to robotic decision-making at the top level. A series of intelligent grasping strategies with rapid slip feedback control and a tactile-visual fusion recognition strategy ensure dexterous robotic grasping and accurate recognition of daily objects, handling various challenging tasks, for instance grabbing a paper cup containing liquid. Furthermore, we showcase a robotic desktop-cleaning task, the robot autonomously accomplishes multi-item sorting and cleaning desktop, demonstrating its promising potential for smart housekeeping.

Abstract New strategies for converting signals between optical and microwave domains could play a pivotal role in advancing both classical and quantum technologies. Traditional approaches to optical-to-microwave transduction typically perturb or destroy the information encoded on intensity of the light field, eliminating the possibility for further processing or distribution of these signals. In this paper, we introduce an optical-to-microwave conversion method that allows for both detection and spectral analysis of microwave photonic signals without degradation of their information content. This functionality is demonstrated using an optomechanical waveguide integrated with a piezoelectric transducer. Efficient electromechanical and optomechanical coupling within this system permits bidirectional optical-to-microwave conversion with a quantum efficiency of up to −54.16 dB. Leveraging the preservation of the optical field envelope in intramodal Brillouin scattering, we demonstrate a multi-channel microwave photonic filter by transmitting an optical signal through a series of electro-optomechanical waveguide segments, each with distinct resonance frequencies. Such electro-optomechanical systems could offer flexible strategies for remote sensing, channelization, and spectrum analysis in microwave photonics.

Abstract Communication by rare, binary spikes is a key factor for the energy efficiency of biological brains. However, it is harder to train biologically-inspired spiking neural networks than artificial neural networks. This is puzzling given that theoretical results provide exact mapping algorithms from artificial to spiking neural networks with time-to-first-spike coding. In this paper we analyze in theory and simulation the learning dynamics of time-to-first-spike-networks and identify a specific instance of the vanishing-or-exploding gradient problem. While two choices of spiking neural network mappings solve this problem at initialization, only the one with a constant slope of the neuron membrane potential at threshold guarantees the equivalence of the training trajectory between spiking and artificial neural networks with rectified linear units. For specific image classification architectures comprising feed-forward dense or convolutional layers, we demonstrate that deep spiking neural network models can be effectively trained from scratch on MNIST and Fashion-MNIST datasets, or fine-tuned on large-scale datasets, such as CIFAR10, CIFAR100 and PLACES365, to achieve the exact same performance as that of artificial neural networks, surpassing previous spiking neural networks. Our approach accomplishes high-performance classification with less than 0.3 spikes per neuron, lending itself for an energy-efficient implementation. We also show that fine-tuning spiking neural networks with our robust gradient descent algorithm enables their optimization for hardware implementations with low latency and resilience to noise and quantization.

Abstract Skillful subseasonal forecasts are crucial for various sectors of society but pose a grand scientific challenge. Recently, machine learning-based weather forecasting models outperform the most successful numerical weather predictions generated by the European Centre for Medium-Range Weather Forecasts (ECMWF), but have not yet surpassed conventional models at subseasonal timescales. This paper introduces FuXi Subseasonal-to-Seasonal (FuXi-S2S), a machine learning model that provides global daily mean forecasts up to 42 days, encompassing five upper-air atmospheric variables at 13 pressure levels and 11 surface variables. FuXi-S2S, trained on 72 years of daily statistics from ECMWF ERA5 reanalysis data, outperforms the ECMWF’s state-of-the-art Subseasonal-to-Seasonal model in ensemble mean and ensemble forecasts for total precipitation and outgoing longwave radiation, notably enhancing global precipitation forecast. The improved performance of FuXi-S2S can be primarily attributed to its superior capability to capture forecast uncertainty and accurately predict the Madden-Julian Oscillation (MJO), extending the skillful MJO prediction from 30 days to 36 days. Moreover, FuXi-S2S not only captures realistic teleconnections associated with the MJO but also emerges as a valuable tool for discovering precursor signals, offering researchers insights and potentially establishing a new paradigm in Earth system science research.

Abstract Legged robots are well-suited for deployment in unstructured environments but require a unique control scheme specific for their design. As controllers optimised in simulation do not transfer well to the real world (the infamous sim-to-real gap), methods enabling quick learning in the real world, without any assumptions on the specific robot model and its dynamics, are necessary. In this paper, we present a generic method based on Central Pattern Generators, that enables the acquisition of basic locomotion skills in parallel, through very few trials. The novelty of our approach, underpinned by a mathematical analysis of the controller model, is to search for good initial states, instead of optimising connection weights. Empirical validation in six different robot morphologies demonstrates that our method enables robots to learn primary locomotion skills in less than 15 minutes in the real world. In the end, we showcase our skills in a targeted locomotion experiment.

Abstract From power plants on land to bridges over the sea, safety-critical built environments require periodic inspections for detecting issues to avoid functional discontinuities of these installations. However, navigation paths in these environments are usually challenging as they often contain difficult-to-access spaces (near-millimetre and submillimetre-high gaps) and multiple domains (solid, liquid and even aerial). In this paper, we address these challenges by developing a class of Thin Soft Robots (TS-Robot: thickness, 1.7 mm) that can access narrow spaces and perform cross-domain multimodal locomotion. We adopted a dual-actuation sandwich structure with a tuneable Poisson’s ratio tensioning mechanism for developing the TS-Robots driven by dielectric elastomers, providing them with two types of gaits (linear and undulating), remarkable output force ( ~ 41 times their weight) and speed (1.16 times Body Length/s and 13.06 times Body Thickness/s). Here, we demonstrated that TS-Robots can crawl, climb, swim and collaborate for transitioning between domains in environments with narrow entries.

Abstract Electrification plays a crucial role in deep decarbonization. However, electrification and power infrastructure can cause mutual challenges. We use nationwide power outage and electric vehicle adoption data in China to provide empirical evidence on how power infrastructure failures can deter electrification. We find that when the number of power outages per district increases by 1 in a given month, the number of new electric vehicles adopted per month decreases by 0.99%. A doubling of power outages in one year on average across the nation can create a depressed adoption rate for up to a decade, implying a decline of more than $ 31.3 million per year in carbon reduction benefits from electric vehicle adoptions. This paper adds to the policy discussion of the costs of increased power outages due to extreme weather and natural disasters, and the urgency for policy to address this issue to facilitate wide adoption of electrification.

Abstract Microstructural changes in conductive materials induced by electric current treatments, such as electromigration and electroplasticity, are critical in semiconductor and metal processing. However, owing to the inevitable thermal effect (Joule heating), the athermal effect on microstructural modifications remains obscure. This paper presents an approach of utilizing pre-micromachined structures, which obstruct current flow but maintain a thermal history similar to that of the matrix, effectively disentangling the thermal and athermal effects. A duplex stainless-steel material is selected to validate the feasibility of this method. Microstructural characterizations show that the athermal effect, especially the electron wind force (EWF), primarily governs the element diffusion and phase transformation in this study. Moreover, many σ phases (Cr-enriched) are precipitated in the micromachined structures, whereas no precipitation occurred in the matrix, suggesting that the directional EWF disrupts the Cr aggregation caused by Joule heating. Furthermore, we present a critical formula for determining the dimensions of micromachined structures of commonly used metallic materials. The proposed method may serve as an effective and powerful tool for unveiling the athermal effect on microstructural alterations.