共找到 20 条结果
Mycobacterium tuberculosis infection is a dynamic continuum. Clinical outcomes reflect complex host-pathogen interactions. Epidemiological and animal studies have suggested influenza coinfection as a risk factor for progression from contained infection to active disease, but human studies have been lacking. Using a whole blood luminescent mycobacterial growth inhibition assay within a human influenza challenge study, we show that influenza infection reduces immunological control of mycobacterial growth. Transcriptome-wide RNA sequencing, cytokine and cellular analyses of subjects' blood before and after influenza infection reveal that innate immune pathways, including type 1 interferon signalling, are activated by influenza but their subsequent responsiveness to mycobacteria is reduced, with multiple genes' responses to BCG lux infection repressed by influenza coinfection. Our data suggest that influenza infection impairs immune mechanisms that contain mycobacterial growth and may be a risk factor for tuberculosis (TB) disease. Influenza vaccination might offer high risk, high prevalence populations protection against TB disease.
Proteins offer a molecular design space to create bespoke ligands for the separation of critical metals like rare earth elements (REs). However, data-intensive approaches to tune metalloprotein selectivity are constrained by the low-throughput nature of existing characterization methods. Here we invented an assay called 'SpyTag-Catcher Immobilization of Lanmodulin for Assaying Metal-Binding Selectivity' (SpyCI-LAMBS) to measure metalloprotein selectivity en masse. This 96-format workflow was used to study the selectivity of 621 lanmodulin (LanM) orthologs for 15 REs, revealing eight distinct selectivity profiles based on sequence-to-function analyses. We discovered >200 LanMs with stronger selectivity against low-value LaIII relative to the prototypical LanM. This includes a LanM that can perform a challenging one-stage separation of PrIII from LaIII with up to >99.9 mol% purity and 83% yield. SpyCI-LAMBS is a powerful tool that can rapidly collect high-fidelity selectivity data to inform metal ion separations and machine-learning-assisted metalloprotein design.
Three-dimensional microfabrication is essential for microfluidics, micromechanical devices, optical components and architected materials, but current methods often trade resolution for speed: layer-by-layer and point-scanning approaches are slow, whereas fast volumetric printing lacks fine features. Here we show a single-exposure holographic lithography method that prints tall, high-resolution polymer microstructures in about 20 s. An inverse-designed phase mask-an optical element computed to shape light in three dimensions-projects a stable intensity pattern through thick photoresist, overcoming the blurring that normally limits deep photolithography. The method produces lattices, Penrose patterns and micromechanical structures with features as small as 6 µm across volumes up to 800 × 800 × 720 µm3 (corresponding to a print rate of 0.36 × 106 voxels/s), achieving aspect ratios above 120:1. The resulting structures guide liquid by capillary action and predictable mechanical behavior under compression. This approach offers a scalable route to complex 3D microstructures for microfluidics, MEMS, optics and architected materials.
The spatiotemporal structure of muscle coordination emerges from the collaboration and competition among cortical, brainstem and spinal pathways onto motor neuron pools, each continuously shaped by task demands, limb position and descending tract integrity. Here, we used galvanic vestibular stimulation (GVS) to investigate whether brainstem vestibular output is disrupted in stroke survivors with right hemiparesis (n = 14) compared with age-matched control subjects (n = 14). We estimated this via intermuscular coherence among neck and arm muscles during both rest and a reaching-like (i.e., reaching) movement with the shoulder in neutral and abducted positions, each with no stimulation, sham stimulation or GVS. Previous work showed that young adults exhibit increased coherence between neck muscles with GVS at rest, as do our control subjects. Surprisingly, we saw the same pattern in stroke survivors only on their paretic side, with reduced coherence in their non-paretic side. During reaching, the paretic side did not show changes in coherence between arm or neck muscles, in comparison to control subjects, with and without GVS, and importantly, even lower coherence with the abducted shoulder. Given that GVS did not exacerbate intermuscular coherence in any muscle pair during the reaching tasks and that coherence was even reduced during shoulder abduction, our findings provide evidence to exclude increased brainstem vestibular output as a dominant contributor to pathological synergies during voluntary movement following stroke. In addition, the decreased coherence between neck muscles of the non-paretic side during GVS at rest suggests changes in ipsilesional brainstem vestibular output. This highlights opportunities to consider unexplored contralesional brainstem-spinal pathways, in addition to downregulated ipsilesional vestibular projections, in neurorehabilitation strategies following stroke.
Pursuing replicability - independent evidence for previous claims - is important for creating generalizable knowledge1,2. Here we attempted replications of 274 claims of positive results from 164 quantitative papers published from 2009 to 2018 in 54 journals in the social and behavioural sciences. Replications were high powered on average to detect the original effect size (median of 99.6%), used original materials when relevant and available, and were peer reviewed in advance through a standardized internal protocol. Replications showed statistically significant results in the original pattern for 151 of 274 claims (55.1% (95% confidence interval (CI) 49.2-60.9%)) and for 80.8 of 164 papers (49.3% (95% CI 43.8-54.7%)), weighed for replicating multiple claims per paper. We observed modest variation in replication rates across disciplines (42.5-63.1%), although some estimates had high uncertainty. The median Pearson's r effect size was 0.25 (95% CI 0.21-0.27) for original studies and 0.10 (95% CI 0.09-0.13) for replication studies, an 82.4% (95% CI 67.8-88.2%) reduction in shared variance. Thirteen methods for evaluating replication success provided estimates ranging from 28.6% to 74.8% (median of 49.3%). Some decline in effect size and significance is expected based on power to detect original effects and regression to the mean because we replicated only positive results. We observe that challenges for replicability extend across social-behavioural sciences, illustrating the importance of identifying conditions that promote or inhibit replicability3,4.
Hepatocellular carcinoma (HCC) screening targets patients with cirrhosis but lacks risk stratification and misses non-cirrhotic cases. Thus, we evaluated whether LIver cancer RIsk Computation (LIRIC) models built from routinely collected electronic health record (EHR) data can predict 3-year HCC risk in general and cirrhosis populations and generalize across diverse US and international cohorts. We conducted a multicenter study using longitudinal EHR data from 64 US and 29 ex-US (Latin America/Asia-Pacific) healthcare organizations. Individuals aged ≥40 years at cohort identification and without previous HCC were included; cases were identified using ICD codes 6-36 months prediagnosis. US models were developed in general and cirrhosis cohorts (46,679 cases; 1,128,202 controls), internally externally validated by site, race/ethnicity, and time, and evaluated in a simulated 'silent' deployment. We benchmarked model-derived risk against population incidence using standardized incidence ratios (SIRs). External validation applied US models to ex-US cohorts; region-specific ex-US models were also trained. Logistic regression and neural networks (a liver risk computation logistic regression model and a liver risk computation neural network model) were compared using AUC, calibration, and geometric mean of overestimation (GMOE) risk ratios. In the US general population, LIRICNN achieved AUC 0.93 (95% CI 0.9218-0.9289) using 46 features. Internal-external AUCs averaged 0.93 across sites and race/ethnicity strata, with good calibration (GMOE 0.89; 95% CI 0.862-0.911). At a screening threshold corresponding to SIR ≈ 31, sensitivity was 48% and specificity was 97%. US-trained models applied to ex-US data yielded lower AUCs (0.84), but region-specific retraining restored performance (AUC 0.94). LIRIC models accurately stratified HCC risk using routine EHR data in both general and cirrhosis populations and could be adapted for international settings via local retraining. These results support LIRIC as a scalable foundation for risk-based HCC surveillance strategies and future prospective implementation studies. Current HCC screening targets patients with cirrhosis but lacks risk stratification and misses non-cirrhotic cases; therefore, we evaluated whether LIRIC models built from routinely collected EHR data can predict 3-year HCC risk in general and cirrhosis populations and generalize across diverse US and international cohorts. These findings are important for hepatologists, primary care clinicians, and health systems because they show that routine EHR data can support accurate and scalable HCC risk stratification across diverse populations. In practice, LIRIC could serve as a foundation for risk-based HCC surveillance strategies and future prospective implementation studies using data already collected in routine care. Given that this was a retrospective modeling study and international performance required local retraining, these findings should be interpreted as supporting future implementation rather than immediate universal adoption.
Depression and suicidality are associated with systematic alterations in cognitive and emotional processing, yet the spatiotemporal neural dynamics underlying these changes during affective task engagement remain poorly characterized. We investigate the time-resolved neural representations of affective semantic processing using multivariate decoding of 64-channel electroencephalography (EEG), while participants (N = 137) perform Sentence Evaluation task using emotionally salient, self-referential statements. Reliable condition-dependent neural discriminability emerges with peak decoding accuracy between 300-600 ms, a time window corresponding to affective semantic evaluation, contextual updating processes, and conflict monitoring. Relative to healthy controls, individuals with depression and suicidal ideation show earlier onset, longer duration, and greater amplitude decoding responses, along with broader cross-temporal generalization and enhanced contributions from frontocentral and parietotemporal components. These results indicate systematic group-level differences in the spatiotemporal dynamics and stability of neural representations during emotionally salient semantic evaluation, contributing to a characterization of neurocognitive processes associated with depression and suicidality.
Frequency combs have revolutionized metrology, ranging and optical clocks1, motivating substantial efforts on the development of chip-scale comb sources2,3. Some on-chip comb sources exist and have been implemented through electro-optic modulation4,5, mode-locked lasers6,7, quantum cascade lasers8-10 or soliton formation by Kerr nonlinearity11,12. However, widespread deployment of on-chip comb sources has remained elusive, as they still require radiofrequency sources, high-Q (high-quality factor) resonators or complex stabilization schemes while facing efficiency challenges. Here, we demonstrate an on-chip frequency comb source based on the integration of a lithium niobate nanophotonic circuit with a semiconductor laser that can alleviate these challenges. We show the formation of temporal topological solitons in an on-chip nanophotonic parametric oscillator with quadratic nonlinearity and low finesse. These solitons, independent of the dispersion regime, consist of phase defects separating two π-out-of-phase continuous wave solutions at the signal frequency, which is half the input pump frequency13,14. We use on-chip cross-correlation for temporal measurements and confirm formation of topological solitons as short as 60 fs around 2 μm, in agreement with a generalized parametrically forced Ginzburg-Landau theory15-17. Moreover, we demonstrate a proof-of-concept turn-key operation of a hybrid-integrated source of topological frequency comb. Topological solitons are potential candidates for the development of integrated comb sources, which are dispersion-sign agnostic and do not require high-Q resonators or high-speed modulators, and can provide access to hard-to-reach spectral regions, including mid-infrared regions18.
Event-based structured light systems have recently emerged as an exciting alternative to frame-based triangulation for 3D measurement of diffuse surfaces, offering high dynamic range and fast capture speed, but at the cost of reduced data quality. Existing event-based and frame-based 3D imaging systems are typically tailored to a single surface reflectance type, diffuse or specular, and therefore struggle with mixed reflectance scenes. In this work, we present an event-based structured light system for fast and accurate 3D imaging of mixed reflectance scenes. Using epipolar constraints on the captured events, we decompose reflections into diffuse, two-bounce specular, and other multi-bounce reflections. Diffuse surfaces are reconstructed via triangulation and repurposed as a virtual screen for deflectometry to evaluate specular surfaces, requiring only a scanning laser and an event camera. Our system achieves motion-robust 3D reconstructions at  < 600 μm depth accuracy and introduces a fast diffuse-only capture mode operating at 250 Hz.
We introduce a decoding framework for correlated errors in quantum LDPC codes under circuit-level noise. Our approach is a graph augmentation and rewiring for inference (GARI) method, which modifies the correlated detector error model by eliminating 4-cycles involving Y-type errors, while preserving the equivalence of the decoding problem. A normalized min-sum decoder with a hybrid serial-layered schedule is applied on the transformed graph, achieving high accuracy with low latency. Performance is further enhanced (on par with XYZ-Relay-BP) through ensemble decoding, where 24 randomized normalized min-sum decoders run in parallel on the transformed graph. For the distance 12 Bivariate Bicycle code the logical error rate of (6.70 ± 1.93) × 10-9 is achieved at a physical error rate of 10-3. Furthermore, preliminary FPGA implementation results show that such high accuracy can be achieved in real time, with a per-round average decoding latency of 273 ns and sub-microsecond latency in 99.99% of the decoding instances.
Time series forecasting is essential across domains such as healthcare, energy, and climate modeling. While models like LSTMs, GRUs, Transformers, and State-Space Models (SSMs) have become widely used, selecting the optimal architecture remains unclear. We propose an automated framework that systematically designs hybrid architectures by combining LSTM, GRU, attention, and SSM modules. Our approach uses multi-objective optimization to explore combinations and orderings of blocks, yielding Pareto-optimal architectures that balance user-defined trade-offs among objectives. A preference function selects the most suitable model for a given application. Moreover, two sampling-based iterative procedures for Pareto-front exploration are introduced, which reduces the total training cost by nearly eightfold. Across four real-world benchmarks, our framework reveals that simple models excel in speed, while hybrid compositions dominate when balancing accuracy and complexity. Our findings challenge the notion of a universally superior neural architecture, emphasizing instead the value of data- and objective-driven design in time series forecasting.
Engineering CRISPR-Cas systems for improved or altered function is critical to both research and therapeutic applications. Unfortunately, most optimization, especially directed evolution in bacterial hosts, fails to capture the functional requirements of the complex mammalian cellular milieu, where activity is usually required. Robust strategies to enable continuous directed evolution of genome-targeting agents directly in human cells remain lacking. Here, we introduce CRISPR-MACE (Mammalian cell-enabled Adenovirus-assisted Continuous Evolution) as a foundational technology to address this need. CRISPR-MACE integrates virus-based continuous evolution with anti-CRISPR-based tunable selection to generate Streptococcus pyogenes Cas9 variants with both increased and decreased DNA binding capacity and nearly 1,000-fold-enhanced resistance to AcrIIA4, the strongest known inhibitor of SpCas9. Notably, across independent evolution campaigns, the same Cas9 gatekeeper mutation reproducibly emerged first, enabling subsequent adaptive steps along two interdependent axes of Cas9 function. In addition to advancing CRISPR technologies, this work establishes key principles and synthetic circuits for continuously evolving CRISPR-Cas systems directly in human cells.
This paper presents the design, fabrication, and characterization of a sub-milliwatt graphene-based micro thermal conductivity detector (µTCD) that utilizes a suspended multilayer graphene (MLG) bridge to sense volatile organic compounds (VOCs) in the gas phase based on their thermal transport properties. The graphene bridge is transferred onto a silicon chip with integrated microchannels using a photolithography-free process. By incorporating microchannel designs, this approach enables precise transfer of suspended MLG dimensions without conventional patterning steps. A key innovation of this work lies in the use of an ultra-low thermal mass suspended graphene architecture, which significantly increases temperature rise per unit input power, thereby enhancing sensitivity per unit power compared to conventional metal-based TCDs. The fabricated µTCD successfully produces chromatograms of multiple VOC species, closely matching those obtained using a standard flame ionization detector (FID). The device demonstrates an estimated limit of detection (LOD) of 190 ppm while consuming an average power of 151 µW under DC operation.
暂无摘要（点击查看详情）
Chiral cavities offer an intriguing way to manipulate material properties by breaking fundamental symmetries. However, only a few chiral cavity implementations exhibiting broken time-reversal symmetry have been demonstrated, with most relying on either strong magnetic fields, circularly polarized Floquet driving, or ultrastrong coupling between cavity modes and matter excitations. Here, we present a one-dimensional terahertz photonic-crystal cavity that exhibits broken time-reversal symmetry. The cavity consists of a silicon wafer sandwiched between InSb wafers. By exploiting the nonreciprocal terahertz response of a magnetoplasma and the low electron effective mass in InSb, a circularly polarized cavity mode at 0.67 THz under a modest magnetic field of 0.3 T, with a quality factor exceeding 50 is realized. Temperature-, magnetic field-, and polarization-dependent measurements and simulations demonstrate the chiral cavity with broken time-reversal symmetry, providing a robust platform for exploring chiral light-matter interactions and vacuum dressed quantum condensed matter in the terahertz regime.
The nociceptin/orphanin FQ peptide (NOP) receptor has emerged as a promising anxiolytic target, as its activation has been shown to reduce anxiety-related behaviors in rodents. However, the mechanisms underlying these effects are not well understood. Here, we investigated the effects of the selective NOP receptor agonist SCH-221510 (0.01-0.1 mg/kg, IM) on behavioral and neural responses to aversive stimuli in squirrel monkeys (n = 3). Subjects underwent Pavlovian fear conditioning, wherein a visual conditioned stimulus (CS) was paired with the presentation of an aversive stimulus. Event-related functional magnetic resonance imaging (fMRI) was conducted in awake subjects to evaluate CS-evoked neural responses. Behavioral and neural responses to the CS were assessed across three experimental phases: pre-conditioning (Pre-C), post-conditioning (Post-C), and Post-C with SCH-221510 administration. In behavioral assessments, CS presentation during Post-C elicited a robust suppression of ongoing operant responding, which was absent during Pre-C and significantly attenuated by SCH-221510 treatment (0.1 mg/kg). fMRI results revealed that, relative to Pre-C, CS presentation during Post-C was associated with increased BOLD activity in brain regions previously implicated in fear processing (e.g., amygdala), expression and regulation (e.g., prefrontal cortex; PFC), as well as sensory integration (e.g., visual cortex). Critically, SCH-221510 (0.1 mg/kg) administration significantly attenuated CS-induced neural activation in these regions. Furthermore, resting-state functional connectivity analysis revealed that SCH-221510 administration decreased connectivity between PFC and amygdala, while enhancing connectivity among PFC subregions. Collectively, these findings suggest that NOP receptor agonism may attenuate conditioned responses to aversive stimuli by modulating functional interactions within a PFC-amygdala circuit.
The supersolid phase is a self-organized state of matter that simultaneously exhibits the crystalline order of a solid and the frictionless flow of a superfluid. Its formation requires the simultaneous breaking of phase and translational symmetries-a stringent condition that makes experimental observation challenging. Here we show that it is possible to achieve a room-temperature supersolid phase by integrating single-crystal halide perovskites with an exciton-polariton nanograting. This architecture supports a hybrid polaritonic bound-state-in-continuum state with a large bandgap (18.2 meV) and two side modes. As the pumping intensity increases, optical parametric oscillation drives the system from a bound-state-in-continuum polariton condensate into the two side modes, forming a self-organized supersolid phase characterized by a striped one-dimensional lattice spanning the condensate. Crucially, single-shot real-space imaging shows stochastic phase selection of the stripe pattern, evidenced by strong suppression of the density modulation on multishot averaging. The observation of supersolidity is further supported by long-range spatiotemporal coherence measured interferometrically and by a non-rigid supersolid lattice. The realization of supersolidity at room temperature in a polaritonic nanograting platform can be useful to control exotic quantum orders and for exploring spontaneous symmetry breaking, quantum coherence and collective excitations in driven quantum materials.
Single-cell foundation models (FMs) pretrained on massive unlabeled scRNA-seq data show strong potential in predicting transcriptional responses to unseen genetic perturbations (e.g. knockouts, variants). However, existing approaches do not effectively transfer pretrained knowledge and overlook the imbalance between perturbation-sensitive and insensitive genes, yielding only marginal improvements over non-pretrained baselines. To address these limitations, we introduce PertAdapt, a framework that unlocks FMs to accurately predict genetic perturbation effects by integrating a plug-in perturbation adapter and an adaptive loss. The adapter employs a gene-similarity-masked attention mechanism to jointly encode perturbation conditions and contextualized representations of unperturbed cells, enabling more effective knowledge transfer. To better capture differential expression patterns, the adaptive loss dynamically reweights perturbation-sensitive genes relative to global transcriptomic signals. Extensive experiments across seven perturbation datasets, including both single- and double-gene settings, demonstrate that PertAdapt consistently outperforms non-pretrained and FM baselines. Moreover, PertAdapt demonstrates strong capacity to model multiplexed gene interactions, to generalize in limited-data regimes, and to maintain robustness across backbone sizes. Code is available at https://github.com/BaiDing1234/PertAdapt.
Engineered programmable RNA sensors have been applied in low-cost diagnostics, endogenous RNA detection, and multi-input genetic circuits. However, designing, producing, and screening high-performance RNA sensors remains time-consuming and labor intensive. Here, we present an automated plasmid assembly pipeline using liquid handling robotics to enable high-throughput construction of plasmids with arbitrary sequences. We compare automated and manual assembly methods using the NGS Hamilton Microlab STAR across two plasmid backbones to evaluate efficiency and reliability. As a proof of concept, we use this modular platform to construct a diverse set of programmable RNA regulators, including toehold switch riboregulators targeting viral RNAs, single-nucleotide-specific programmable riboregulators for discrimination of SARS-CoV-2 spike gene mutations, and metal-responsive riboswitches. In total, we construct 174 plasmids and test the designed methods by comparing both manual and automated assembly. We further demonstrate that the assembled toehold switch plasmids are functional in both bacterial and cell-free expression systems.