ETH Zurich

Dynamic sparsity is intrinsic to biological computing and is key to its extreme power efficiency. Edge computing systems can improve their energy efficiency and reduce response latency by exploiting this neuromorphic principle. The neuromorphic approach for the extraction of acoustic features replaces conventional ADC and DSP with biological cochlea-inspired filters and event generators implemented in mixed-signal circuits. The resulting sparse feature events drive inference in dynamic-sparsity-aware neural network accelerators to reduce computational load and memory access. The demonstration of edge keyword spotting shows the dynamic savings in power. Exploiting dynamic sparsity at all levels will be the next step toward the design of intelligent devices for the edge.

Accurate tire modeling is crucial for optimizing autonomous racing vehicles, as State-of-the-Art (SotA) modelbased techniques rely on precise knowledge of the vehicle's parameters, yet system identification in dynamic racing conditions is challenging due to varying track and tire conditions. Traditional methods require extensive operational ranges, often impractical in racing scenarios. Machine Learning (ML)-based methods, while improving performance, struggle with generalization and depend on accurate initialization. This paper introduces a novel ontrack system identification algorithm, incorporating a NN for error correction, which is then employed for traditional system identification with virtually generated data. Crucially, the process is iteratively reapplied, with tire parameters updated at each cycle, leading to notable improvements in accuracy in tests on a scaled vehicle. Experiments show that it is possible to learn a tire model without prior knowledge with only 30 seconds of driving data, and 3 seconds of training time. This method demonstrates greater one-step prediction accuracy than the baseline Nonlinear Least Squares (NLS) method under noisy conditions, achieving a 3.3x lower Root Mean Square Error (RMSE), and yields tire models with comparable accuracy to traditional steady-state system identification. Furthermore, unlike steady-state methods requiring large spaces and specific experimental setups, the proposed approach identifies tire parameters directly on a race track in dynamic racing environments.

Persistent multiyear drought (MYD) events pose a growing threat to nature and humans in a changing climate. We identified and inventoried global MYDs by detecting spatiotemporally contiguous climatic anomalies, showing that MYDs have become drier, hotter, and led to increasingly diminished vegetation greenness. The global terrestrial land affected by MYDs has increased at a rate of 49,279 ± 14,771 square kilometers per year from 1980 to 2018. Temperate grasslands have exhibited the greatest declines in vegetation greenness during MYDs, whereas boreal and tropical forests have had comparably minor responses. With MYDs becoming more common, this global quantitative inventory of the occurrence, severity, trend, and impact of MYDs provides an important benchmark for facilitating more effective and collaborative preparedness toward mitigation of and adaptation to such extreme events.

Incorporation of animal-based foods into early hominin diets has been hypothesized to be a major catalyst of many important evolutionary events, including brain expansion. However, direct evidence of the onset and evolution of animal resource consumption in hominins remains elusive. The nitrogen-15 to nitrogen-14 ratio of collagen provides trophic information about individuals in modern and geologically recent ecosystems (<200,000 years ago), but diagenetic loss of this organic matter precludes studies of greater age. By contrast, nitrogen in tooth enamel is preserved for millions of years. We report enamel-bound organic nitrogen and carbonate carbon isotope measurements of Sterkfontein Member 4 mammalian fauna, including seven Australopithecus specimens. Our results suggest a variable but plant-based diet (largely C 3 ) for these hominins. Therefore, we argue that Australopithecus at Sterkfontein did not engage in regular mammalian meat consumption.

Conductive metal–organic frameworks (MOFs) are crystalline, intrinsically porous materials that combine remarkable electrical conductivity with exceptional structural and chemical versatility. This rare combination makes these materials highly suitable for a wide range of energy‐related applications. However, the electrical conductivity in MOF‐based devices is often limited by the presence of different types of structural disorder. Here, the electrical transport characteristics of high quality Ni3(HITP)2 nanometer‐thin films are reported. These findings reveal a tenfold difference in conductivity between the micro‐ and nano‐scale, attributed to poor electrical connection among a limited number of crystalline grains. Average in‐plane conductivity values at the micro‐ (σIP,micro = 0.7 ± 0.3 S cm⁻¹) and nano‐ (σIP,nano = 6 ± 3 S cm⁻¹) scales is determined, and the value of the inter‐grain resistance, Rinter‐grain = 40 kΩ is found. Using a 2D resistor network model with a 40 kΩ base resistance and scattered higher resistances, surface potential maps of in‐operando MOF‐based electrical devices are successfully reproduced. Additionally, a structure–property relationship that links the density and spatial distribution of electrical failures in inter‐grain connections to the observed micro‐scale conductivity in MOF thin films is established.

Autoimmune hepatitis (AIH) is a rare chronic inflammatory liver disease characterized by the presence of autoantibodies, including those targeting O-phosphoseryl-tRNA:selenocysteine-tRNA synthase (SepSecS), also known as soluble liver antigen (SLA). Anti-SepSecS antibodies have been associated with a more severe phenotype, suggesting a key role for the SepSecS autoantigen in AIH. To analyze the immune response to SepSecS in patients with AIH at the clonal level, we combined sensitive high-throughput screening assays with the isolation of monoclonal antibodies (mAbs) and T cell clones. The anti-SepSecS mAbs isolated were primarily IgG1, affinity-matured compared with their germline versions, and recognized at least 3 nonoverlapping epitopes. SepSecS-specific CD4⁺ T cell clones were found in patients with AIH who were anti-SLA-positive and anti-SLA-negative,and, to a lesser extent, in patients with non-AIH liver diseases and in healthy individuals. SepSecS-specific T cell clones from patients with AIH produced IFN-γ, IL-4, and IL-10, targeted multiple SepSecS epitopes, and, in one patient, were clonally expanded in both blood and liver biopsy. Finally, SepSecS-specific B cell clones, but not those of unrelated specificities, were able to present soluble SepSecS to specific T cells. Collectively, our study provides the first detailed analysis of B and T cell repertoires targeting SepSecS in patients with AIH, offering a rationale for improved targeted therapies.

This study seeks to explore two new km-scale regional climate simulations prepared through the European Climate Prediction project over the Madeira and Canary islands, which are Portuguese and Spanish archipelagos located in the North Atlantic, off the African coast. The simulations are based on two models using different modelling approaches: COSMO-CLM with a three-step nesting at 50, 25 and 3 km grid spacing using a time-slice approach driven by a global climate model, and COSMO-crCLIM with a two-step nesting at 12 and 1 km grid spacing using the pseudo-global warming approach, where the current-day simulations are driven by ERA-Interim reanalysis. Although the modelling approaches are different, several findings are highlighted: (1) the use of km-scale simulation is essential to properly represent temperature and precipitation mean and extremes over small islands that are characterized by complex topography; (2) the projected changes in temperature and precipitation mean and extremes are qualitatively similar in all seasons except autumn; (3) the differences in the autumn projections are shown to be due to the large-scale driving conditions. Small islands, such as the Canary and Madeira ones, are often neglected by large modelling initiatives, so the presented simulations contribute to filling this gap for local policy makers, stakeholders and climate services. The encouraging results highlight the need for further coordinated km-scale projections.

Recently, Lin and Shinder constructed non-trivial homomorphisms from Cremona groups of rank $>2$ to $\mathbb {Z}$ using motivic techniques. In this short note we propose an alternative perspective from median geometry on their theorem.

Semiconductor‐superconductor hybrid materials are used as a platform to realize Andreev bound states, which hold great promise for quantum applications. These states require transparent interfaces between the semiconductor and superconductor, which are typically realized by in‐situ deposition of an Al superconducting layer. Here a hybrid material is presented, based on an InAs 2D electron gas (2DEG) combined with in‐situ deposited Nb and NbTi superconductors, which offer a larger operating range in temperature and magnetic field due to their larger superconducting gap. The inherent difficulty associated with the formation of an amorphous interface between III‐V semiconductors and Nb‐based superconductors is addressed by introducing a 7 nm Al interlayer. The Al interlayer provides an epitaxial connection between an in‐situ magnetron sputtered Nb or NbTi thin film and a shallow InAs 2DEG. This metal‐to‐metal epitaxy is achieved by optimization of the material stack and results in an induced superconducting gap of approximately 1 meV, determined from transport measurements of superconductor‐semiconductor Josephson junctions. This induced gap is approximately five times larger than the values reported for Al‐based hybrid materials and indicates the formation of highly‐transparent interfaces that are required in high‐quality hybrid material platforms.

In situ monitoring is essential for catalytic process design, offering real‐time insights into active structures and reactive intermediates. Electron paramagnetic resonance (EPR) spectroscopy excels at probing geometric and electronic properties of paramagnetic species during reactions. Yet, state‐of‐the‐art liquid‐phase EPR methods, like flat cells, require custom resonators, consume large amounts of reagents, and are unsuited for tracking initial kinetics or use with solid catalysts. To overcome these limitations, a droplet‐based microfluidics platform is introduced for real‐time EPR monitoring of liquid‐phase catalytic reactions. By encapsulating solid and dissolved species within nanoliter droplets, this approach enables precise control over mass transport, reduces reagent consumption, and maintains uniform residence times irrespective of acquisition duration, permitting precise analysis of each spectral component under identical conditions. The platform's compatibility with standard resonators facilitates straightforward integration into any EPR spectrometer. Its versatility is demonstrated by monitoring dynamic ligand exchange processes, key for activating homogeneous catalysts, and tracking redox and radical kinetics in ascorbic acid oxidation by Cu(II) catalysts. Importantly, this method captures both supported and dissolved transition metal species, offering comprehensive insights into catalyst deactivation via metal leaching. This microfluidic approach sets a new standard for liquid‐phase in situ EPR measurements, advancing studies of homogeneous and heterogeneous catalytic systems.

Increasing soil salinity causes significant crop losses globally; therefore, understanding plant responses to salt (sodium) stress is of high importance. Plants avoid sodium toxicity through subcellular compartmentation by intricate processes involving a high level of elemental interdependence. Current technologies to visualize sodium, in particular, together with other elements, are either indirect or lack in resolution. Here we used the newly developed cryo nanoscale secondary ion mass spectrometry ion microprobe¹, which allows high-resolution elemental imaging of cryo-preserved samples and reveals the subcellular distributions of key macronutrients and micronutrients in root meristem cells of Arabidopsis and rice. We found an unexpected, concentration-dependent change in sodium distribution, switching from sodium accumulation in the cell walls at low external sodium concentrations to vacuolar accumulation at stressful concentrations. We conclude that, in root meristems, a key function of the NHX family sodium/proton antiporter SALT OVERLY SENSITIVE 1 (also known as Na⁺/H⁺ exchanger 7; SOS1/NHX7) is to sequester sodium into vacuoles, rather than extrusion of sodium into the extracellular space. This is corroborated by the use of new genomic, complementing fluorescently tagged SOS1 variants. We show that, in addition to the plasma membrane, SOS1 strongly accumulates at late endosome/prevacuoles as well as vacuoles, supporting a role of SOS1 in vacuolar sodium sequestration.

Noncoding satellite DNA repeats are abundant at the pericentromeric heterochromatin of eukaryotic chromosomes. During interphase, sequence-specific DNA-binding proteins cluster these repeats from multiple chromosomes into nuclear foci known as chromocenters. Despite the pivotal role of chromocenters in cellular processes like genome encapsulation and gene repression, the associated proteins remain incompletely characterized. Here, we use 2 satellite DNA-binding proteins, D1 and Prod, as baits to characterize the chromocenter-associated proteome in Drosophila embryos, ovaries, and testes through quantitative mass spectrometry. We identify D1- and Prod-associated proteins, including known heterochromatin proteins as well as proteins previously unlinked to satellite DNA or chromocenters, thereby laying the foundation for a comprehensive understanding of cellular functions enabled by satellite DNA repeats and their associated proteins. Interestingly, we find that multiple components of the transposon-silencing piRNA pathway are associated with D1 and Prod in embryos. Using genetics, transcriptomics, and small RNA profiling, we show that flies lacking D1 during embryogenesis exhibit transposon expression and gonadal atrophy as adults. We further demonstrate that this gonadal atrophy can be rescued by mutating the checkpoint kinase, Chk2, which mediates germ cell arrest in response to transposon mobilization. Thus, we reveal that a satellite DNA-binding protein functions during embryogenesis to silence transposons, in a manner that is heritable across later stages of development.

Traditional fraud detection approaches often use linking entities, such as device, email, and address, to identify fraudulent transactions and users. However, as fraud methods continue to evolve and escalate, the fraudsters can fabricate the involved entities and thus hide their real intent. To make fraud detection more robust, we incorporate user behaviors in the pipeline and consider biometric characteristics that are difficult to forge. In this work, we conduct a detailed study of how user behavior data can help identify and prevent fraudulent activity in e-commerce. We present Multi-Modal Behavioral Transformer (MMBT), where we combine both inner-page behavioral data, such as mouse trajectory, and inter-page behavioral data, such as page view sequences. We propose to construct mouse trajectory data as an image, treat each mouse position as a pixel in the image, convert the image into small patches, and hence transform the mouse trajectory into patch index sequences. Our experimental results on real-word data show that MMBT significantly outperforms baselines — the precision@recall = 0.1 increases by up to 7%. In addition, we have built an online pipeline to operationalize our model. In production, the 99th percentile latency is maintained below 500 milliseconds, allowing the platform to initiate rapid response measures and prevent potential losses.

The two‐fold reduction of tetrabenzo[a,c,e,g]cyclooctatetraene (TBCOT, or tetraphenylene, 1) with K, Rb, and Cs metals reveals a distinctive core transformation pathway: a newly formed C−C bond converts the central eight‐membered ring into a twisted core with two fused five‐membered rings. This C−C bond of 1.589(3)–1.606(6) Å falls into a single σ‐bond range and generates two perpendicular π‐surfaces with dihedral angles of 110.3(9)°–117.4(1)° in the 1TR²⁻ dianions. As a result, the highly contorted 1TR²⁻ ligand exhibits a “butterfly” shape and could provide different coordination sites for metal‐ion binding. The K‐induced reduction of 1 in THF affords a polymeric product with low solubility, namely [{K⁺(THF)}2(1TR²⁻)] (K2‐1TR²⁻). The use of a secondary ligand facilitates the isolation of discrete complexes with heavy alkali metals, [Rb⁺(18‐crown‐6)]2[1TR²⁻] (Rb2‐1TR²⁻) and [Cs⁺(18‐crown‐6)]2[1TR²⁻] (Cs2‐1TR²⁻). Both internal and external coordination are observed in K2‐1TR²⁻, while the bulky 18‐crown‐6 ligand only allows external metal binding in Rb2‐1TR²⁻ and Cs2‐1TR²⁻. The reversibility of the two‐fold reduction and bond rearrangement is demonstrated by NMR spectroscopy. Computational analysis shows that the heavier alkali metals enable effective charge transfer from the 1TR²⁻TBCOT dianion, however, the aromaticity of the polycyclic ligand remains largely unaffected.

In some fields of artificial intelligence, machine learning and statistics, the validation of new methods and algorithms is often hindered by the scarcity of suitable real-world datasets. Researchers must often turn to simulated data, which yields limited information about the applicability of the proposed methods to real problems. As a step forward, we have constructed two devices that allow us to quickly and inexpensively produce large datasets from non-trivial but well-understood physical systems. The devices, which we call causal chambers, are computer-controlled laboratories that allow us to manipulate and measure an array of variables from these physical systems, providing a rich testbed for algorithms from a variety of fields. We illustrate potential applications through a series of case studies in fields such as causal discovery, out-of-distribution generalization, change point detection, independent component analysis and symbolic regression. For applications to causal inference, the chambers allow us to carefully perform interventions. We also provide and empirically validate a causal model of each chamber, which can be used as ground truth for different tasks. The hardware and software are made open source, and the datasets are publicly available at causalchamber.org or through the Python package causalchamber.

The redox signaling network in mammals has garnered enormous interest and taken on major biological significance in recent years as the scope of NADPH oxidases (NOXs) as regulators of physiological signaling and cellular degeneration has grown exponentially. All NOX subtypes have in common the capacity to generate reactive oxygen species (ROS) superoxide anion (O 2 .- ) and/or hydrogen peroxide (H 2 O 2 ). A baseline, normal level of ROS formation supports a wide range of processes under physiological conditions. A disruption in redox balance caused by either the suppression or "super" induction of NOX off balance with antioxidant systems is associated with myriad diseases and cell/tissue damage. Over the past two to three decades sour understanding of NOXs has progressed from almost entirely a phagocyte-, antimicrobial-centered perspective to that of a family of enzymes that is vital to broad cellular function and organismal homeostasis. It is becoming increasingly evident that highly regulated, targeted oxidative protein modifications are elicited in a spatiotemporal manner and initiated at cell membranes in humans by seven NOX isoforms (NOXs 1, 2, 3, 4, 5 and DUOXs 1 & 2). In a sense, this renders NOX-ROS signaling akin to that of other second messenger systems involving localized Ca ²⁺ dynamics and tyrosine kinase transactivation. Accordingly, the study of ROS compartmentalization in subcellular organelles has been shown to be crucial to elucidating their role in cell phenotype modulation under physiological and pathophysiological conditions. The NOXs are as distinct in their distribution and activation as they are in their cellular functions, ranging from host defense, second messenger PTMs to transcriptional, epigenetic and (de)differentiating effects. The review integrates past knowledge in the field with new focus areas on the leading-edge of NOX-centered ROS signaling including how a new wave of structural information provides insights for NOX biology and targeted therapies.

INTRODUCTION Transcranial pulse stimulation (TPS) is increasingly being investigated as a promising potential treatment for Alzheimer's disease (AD). Although the safety and preliminary clinical efficacy of TPS short pulses have been supported by neuropsychological scores in treated AD patients, its fundamental mechanisms are uncharted. METHODS Herein, we used a multi‐modal preclinical imaging platform combining real‐time volumetric optoacoustic tomography, contrast‐enhanced magnetic resonance imaging, and ex vivo immunofluorescence to comprehensively analyze structural and hemodynamic effects induced by TPS. Cohorts of healthy and AD transgenic mice were imaged during and after TPS exposure at various per‐pulse energy levels. RESULTS TPS enhanced the microvascular network, whereas the blood–brain barrier remained intact following the procedure. Notably, higher pulse energies were necessary to induce hemodynamic changes in AD mice, arguably due to their impacted vessels. DISCUSSION These findings shed light on cerebrovascular dynamics induced by TPS treatment, and hence are expected to assist improving safety and therapeutic outcomes. Highlights ·Transcranial pulse stimulation (TPS) facilitates transcranial wave propagation using short pulses to avoid tissue heating. ·Preclinical multi‐modal imaging combines real‐time volumetric optoacoustic (OA) tomography, contrast‐enhanced magnetic resonance imaging (CE‐MRI), and ex vivo immunofluorescence to comprehensively analyze structural and hemodynamic effects induced by TPS. ·Blood volume enhancement in microvascular networks was reproducibly observed with real‐time OA imaging during TPS stimulation. ·CE‐MRI and gross pathology further confirmed that the brain architecture was maintained intact without blood–brain barrier (BBB) opening after TPS exposure, thus validating the safety of the procedure. ·Higher pulse energies were necessary to induce hemodynamic changes in AD compared to wild‐type animals, arguably due to their pathological vessels.

Climate change is causing many species' ranges to shift upslope to higher elevations as species track their climatic requirements. However, many species have not shifted in pace with recent warming (i.e. ‘range stasis'), possibly due to demographic lags or microclimatic buffering. The ‘lagged‐response hypothesis' posits that range stasis disguises an underlying climatic sensitivity if range shifts lag the velocity of climate change due to slow colonization (i.e. colonization credits) or mortality (i.e. extinction debt). Alternatively, the ‘microclimatic buffering hypothesis' proposes that small‐scale variation within the landscape, such as canopy cover, creates patches of suitable habitat within otherwise unsuitable macroclimates that create climate refugia and buffer range contractions. We simultaneously test both hypotheses by combining a large seed addition experiment of 25 plant species across macro‐ and micro‐scale climate gradients with adult occurrence records to compare patterns of seedling recruitment relative to adult ranges and microclimate in the North Cascades, USA. Despite high species‐to‐species variability in recruitment, community‐level patterns monitored for five years supported the lagged response hypothesis, with a mismatch between where seedlings recruit versus adults occur. On average, the seedling recruitment optimum shifted from the adult climatic range centre to historically cooler, wetter regions and many species recruited beyond their cold (e.g. leading) range edge. Meanwhile, successful recruitment occurred at warm and dry edges, despite recent climate change, suggesting that macroclimatic effects on recruitment do not drive trailing range dynamics. We did not detect evidence of microclimatic buffering due to canopy cover in recruitment patterns. Combined, our results suggest apparent range stasis in our system is a lagged response to climate change at the cool ends of species ranges, with range expansions likely to occur slowly or in a punctuated fashion.