The separation of Bacillus thuringiensis (Bt) spores and crystals is often compromised by dense aggregates, particularly in Bt subsp. israelensis. This study evaluates (sonication), chemical (SDS), and biological (lipopeptide) treatments to improve aggregate dissociation in BUPM98, using AR23 as a reference and Lip (Bt subsp. kurstaki) as a naturally dissociated control. Initial hexane-based purification highlighted the impact of aggregation, with significantly lower recovery rates for israelensis compared to kurstaki. Electron microscopic analysis revealed tightly integrated spore crystal complexes in Bt israelensis and independent bipyramidal crystals in Lip. While physical sonication and chemical SDS treatments improved dissociation, biological treatment with lipopeptides yielded the most significant results. This approach drastically increased viable spore counts by up to 8000-fold and enhanced crystal recovery to 60%, offering a superior and more efficient biosurfactant-based alternative to traditional methods. PCR analysis identified the kurstakin synthetase gene exclusively in the naturally dissociated kurstaki strain, suggesting a native role for these biosurfactants in preventing aggregation. To our knowledge, this is the first demonstration that lipopeptides can effectively dissociate Bacillus thuringiensis spore crystal aggregates, offering an efficient biosurfactant-based alternative to physical or chemical methods.
Molecular communication is a model of information transmission where the signal is transmitted by information-carrying molecules through their physical transport from a transmitter to a receiver through a communication channel. Prior efforts have identified suitable "information molecules" whose efficacy for signal transmission has been studied extensively in diffusive channels (DC). Although easy to implement, DCs are inefficient for distances longer than tens of nanometers. In contrast, molecular motor-driven nonequilibrium or active transport can drastically increase the range of communication and may permit efficient communication up to tens of micrometers. In this paper, we investigate how active transport influences the efficacy of molecular communication, quantified by the mutual information between transmitted and received signals. We consider two specific scenarios: (a) active transport through relays and (b) active transport through a mixture of active and diffusing particles. In each case, we discuss the efficacy of the communication channel and discuss their potential pitfalls.
Wireless Power Transfer-based electric vehicles (EVs) emerge as a promising technology that can drastically reduce carbon emissions worldwide. Conventional ICE-based transportation has led to a significant increase in greenhouse gas emissions and exacerbated climate change. On the contrary, WPT-based EVs can offer extended range, contactless charging, enhanced safety, and reduced vehicle costs. However, the complex design of WPT is vulnerable to power converter failure and coil misalignment. These faults can significantly deteriorate the performance of the WPT-based EVs. A power converter failure can degrade WPT performance, and this paper introduces redundant switches to ensure smooth vehicle operation. This paper presents a Physics-informed Neural Network (PINN) that incorporates physical laws to observe signals from power converters and control the Active Fault-Tolerant Controller (AFTC). The AFTCS consists of the FDI block, which is responsible for detecting and isolating faults in real time. The PINN-based observer system monitors the signal in real time. The results clearly illustrate that the proposed framework provides accurate current tracking and ensures the residual signal remains in the threshold even in parameter variations and achieves a reduced settling time of 0.22s. Furthermore, the steady-state error is reduced to 0.3 A, outperforming the other control methods and ensuring the stability of the system in the event of faults. Thus, a PINN-based FTC is introduced to ensure the resilience and reliable operation of WPT-based EVs in case of power converter failure and coil misalignment.
Severe dysarthria and global aphasia drastically reduce speech intelligibility, confining communication to familiar partners. Automatic speech recognition (ASR) systems may show limited performance when processing such atypical speech. To determine whether a speaker-dependent Voice-Input Voice-Output Communication Aid (VIVOCA) embedded in the CapisciAMe app can decode the speech of a person with severe dysarthria and aphasia more accurately than rehabilitation professionals human listeners (RPHL). We conducted a single-case proof-of-concept study. A 34-year-old woman, 15 years post-stroke, recorded 1,120 utterances of 13 target-words across five prompting modalities. A compact convolutional neural network (cnn-trad-fpool3) was trained on these samples and evaluated on an independent set of 936 utterances. Intelligibility was benchmarked against 12 RPHL familiar with the patient. The primary outcome was word-level accuracy. The tailored ASR achieved 72.65 % accuracy, outperforming familiar RPHL (mean = 56.75 %, SD = 12.91). A personalized ASR system can exceed the intelligibility of human listeners for profoundly disordered speech, supporting its use as an assistive communication technology.
Wheat bran is an important source of dietary fiber in whole wheat products, yet the high content of insoluble dietary fiber (IDF) drastically impairs dough rheology and bread quality. In this study, solid-state fermentation was applied to wheat bran using ten tea-derived starters. All starters promoted IDF degradation and induced a transient increase in soluble dietary fiber. Among them, QD and FY showed superior modification efficiency, remarkably reducing the crystallinity of cellulose. Consequently, the dough prepared with fermented bran exhibited reduced viscoelastic stiffness and improved flow behavior, while the corresponding breads showed increased specific volume and reduced crumb hardness. Microbiome analysis of QD and FY showed that fermentation markedly reduced bacterial and fungal diversity, selectively enriching the genera of Bacillus and Aspergillus with cellulolytic and xylanolytic potential. Overall, this study provided new insights for modifying wheat bran composition and structure, thereby improving the processing quality of whole wheat bread. The online version contains supplementary material available at 10.1007/s10068-026-02189-7.
Deep learning-based segmentation has become essential in computer-aided dental diagnosis and treatment planning. However, these models remain highly vulnerable to adversarial perturbations, like small and imperceptible changes in input images, which can drastically alter segmentation outputs and compromise clinical reliability. In this work, we present the first systematic study on adversarial vulnerability and robustness of deep learning models for panoramic dental X-ray segmentation. We curated a dataset of 995 panoramic images by combining 361 expert-annotated radiographs with 634 refined masks from the DENTEX 2023 challenge. Under identical training conditions, initially, we benchmarked 11 unique model variants, including five core architectures (Attention UNet, SegNet, Trans UNet, Vanilla UNet, and UNet++) and their corresponding ablations on training and preprocessing techniques. UNet++ emerged as the most practical backbone (clean IoU [Formula: see text], Dice [Formula: see text]) and subjected to a suite of white-box attacks with FGSM, I-FGSM, PGD, and DeepFool across perturbation ([Formula: see text]). Our results reveal that even minimal perturbations caused large performance drops, such as at [Formula: see text], IoU collapsed to 23.5% (0.851 to 0.649). To mitigate this fragility, we implemented a customized multi-attack adversarial defense strategy to ensure the model's robustness, which preserved a modest clean-accuracy trade-off by increasing 14.9% (IoU 0.649 to 0.798) at [Formula: see text] and 12.5% at [Formula: see text]. Our qualitative and quantitative analyses demonstrate that the defended model produces more stable and anatomically consistent masks under attack and set the benchmark of adversarial robustness in dental image segmentation as an effective defense strategy for safety-critical clinical deployment.
This paper addresses the swing attenuation problem in quadrotor slung-load systems, which is challenging due to nonlinearity, underactuation, and strong dynamic coupling. A robust partitioned control strategy is proposed to handle model uncertainties, external disturbances, and load variations. By establishing the differential flatness of the system with load position and quadrotor yaw angle as flat outputs, the dynamics are partitioned into a fully actuated subsystem (FAS) and an underactuated subsystem (UAS). For the FAS, a novel adaptive nonsingular fast integral terminal sliding mode (AFITSM) controller is developed. It ensures finite-time convergence and enhances robustness through online estimation of lumped uncertainties. For the UAS, an overall sliding mode control (OSMC) strategy is formulated to coordinate lateral motion for swing damping. Comprehensive validation studies demonstrate that the proposed AFITSM-OSMC reduces tracking RMSE by 89%-99% and swing angle by 87%-93% compared to benchmark methods, while guaranteeing finite-time convergence within 1.7s under various uncertainties and disturbances. Under dynamic load variations, it limits maximum swing to 7.2∘ and achieves rapid stabilization, whereas benchmarks fail to settle. Moreover, control chattering is drastically suppressed, and peak torque rates are reduced by over 85%. The scheme offers superior tracking precision, rapid swing attenuation, robustness, and practical deployability.
Heterogenized molecular transition metal complexes, especially silver-based ones, have shown to be highly efficient as catalysts for electrochemical CO2 reduction (eCO2R). Herein we present the application of a silver dithiacyclam polymer (dithiacyclam = 1,8-dithia-4,11-diazacyclotetradecane, Ag(dithiacyclam)) in homogeneous as well as heterogeneous eCO2R. Although the complex does not exhibit a noteworthy activity in solution, changing the catalyst environment through heterogenization onto a gas diffusion electrode (GDE) and subsequent application in a zero-gap electrolyzer (ZGE) drastically boosts its catalytic performance. At a current density of 50 mA cm-2 a remarkable FECO of 97% is reached. Moreover, we highlight how optimization of the GDE fabrication via ink engineering including the choice of dispersion solvent results in a FECO up to 90% at an elevated current density of 300 mA cm-2. Even at more application oriented current densities of 500 mA cm-2 the eCO2R outcompetes the competing hydrogen evolution reaction, achieving a FECO of 55%. Although signs of catalyst transformation into silver particles are observed in post-mortem analysis, these particles show higher activity than commercially available silver nanoparticles, thus highlighting that molecular systems can be very promising catalyst precursors for efficient eCO2R.
Elevated intracellular glutathione, intratumoral hypoxia, and excessive polyamines severely compromise the efficacy of radiotherapy (RT) by attenuating radiation damage and maintaining an immunosuppressive tumor microenvironment (ITME). Strategies that concurrently address these obstacles are urgently needed to enhance RT outcomes. Herein, we fabricated a biomimetic nanoplatform (PCuP) via the self-assembly of copper ions and piceatannol into CuP nanometal-polyphenol cores, followed by surface camouflage with platelet membranes (PM). Owing to the platelet membrane coating, PCuP preferentially accumulates in tumor tissues with disrupted vasculature. At the tumor site, PCuP catalyzes hydrogen peroxide to produce oxygen, relieving hypoxia and sensitizing the tumors to RT. Simultaneously, it generates reactive oxygen species (ROS) that cooperate with RT to induce immunogenic cell death (ICD). Furthermore, PCuP inhibits arginase 2 (Arg2) to deplete polyamines, suppressing tumor DNA repair and remodeling the ITME. In vivo experiments using oral squamous cell carcinoma (OSCC) models demonstrated that PCuP promotes dendritic cell maturation, activates systemic antitumor immunity, and drastically reduces post-RT tumor recurrence. This study presents the first nanoplatform integrating polyamine depletion with radioimmunotherapy, offering a promising strategy to design advanced radiosensitizers for improving RT efficacy and lowering tumor recurrence.
A scCO2 drying/lixiviation process allows self-standing inorganic macrocelullar aerogels to be produced within 1 h. The applied depressurization rate permits the macrocellular internal throats to be tuned with native hydrodynamic instability. scCO2 treatment advantageously addresses lixiviation of the tensioactive molecules employed as the templating agent. Final calcined aero-Si(HIPE) materials, at the optimum temperature of 450 °C, afford a specific surface area of up to 1400 m2 g-1. The lower treatment temperature and lower tension-active contents dramatically reduce the carbon footprint penalty, compared with that of traditional xero-Si(HIPE). Further annealing of these aero-Si(HIPE) foams at 1200 °C drastically decreases the specific surface area while maintaining both the open macroposity and the self-standing character. Concomitantly, severe shrinkage and densification of the monoliths occur upon cristobalite crystallization. For the sake of making this generic endeavor, the scCO2 drying/lixiviation approach has also been extended to MUB-110, MUB-200, and MUB-300 co-oxide(HIPE) catalysts describing the same scenario of fostering the specific surface area, maintaining the monolith-type character and associated macroporous open tortuosity, and minimizing the carbon footprint penalty.
Human T-lymphotropic virus (HTLV-1), a delta retrovirus, encodes for accessory proteins in addition to structural and enzymatic proteins. p30, an HTLV-1 accessory protein, contributes to viral persistence and pathogenesis. p30 is a multifunctional protein and its intrinsically disordered protein (IDP) nature enables interactions with different host proteins. The phosphorylation drastically impacts the structure and function of IDPs. In this study, we analyzed the intrinsic disorderness of p30 protein under different levels of phosphorylation, dynamics and the interaction with the regulatory protein Rex. Through MD studies, the effect of phosphorylation on the IDP nature of p30 were examined. The analysis revealed that the p30 with phosphorylation at 10 predicted sites (p30-10site(-P)), exhibited more stable behavior and higher compactness compared to p30 with phosphorylation at 3 experimentally identified sites (p30-3site(-P)) and apo-protein. Rex is a regulatory protein encoded by HTLV-1. The interaction between p30 and Rex modulates the balance between viral replication and latency. Due to the presence of ID regions in p30 and Rex, a significant percentage of loops were observed in their secondary structure. The stability of p30 in the p30-Rex complex is also influenced by its binding partners (Rex protein), as the binding sites are located within the ID regions of p30. It can be concluded that the phosphorylation level and Rex modulate the conformational ensemble of p30. The limited phosphorylation together with Rex interaction may contribute to the structural stability of p30 which is required for its multifunctional role in HTLV-1 infection.
The worldwide climate is thought to be drastically changing as a result of the global temperatures, a phenomenon known as "global warming". Thermal stress is a crucial obstacle facing buffalo cyclicity. Investigation of the molecular regulation concerning proliferation and apoptosis of corpus luteum (CL) is not fully comprehended in buffaloes. We aimed to (1) study mRNA expression of candidate genes related to proliferation (PGR, AGTR1, and LHCGR) and apoptosis (TNFα, BAX, FASLG, CASP3, AGTR2 and PTGS2), (2) explore effect of thermal stress on the expression of HSP70, NOS1, NOS2 mRNAs, NO and SOD concentrations in CL homogenate during different stages of CL. For this, ovaries (n = 70) were collected in pairs from buffaloes during cold and hot seasons. According to morphology of CL, samples were divided into: early, mid, and late. For RNA isolation, NO and SOD concentrations, small sections from CL stages were frozen in - 80 °C. The results showed that PGR, AGTR2, TNFα, BAX, cALP2beta and PTGS2 mRNAs decreased (P < 0.001) at different stages of CL at hot season. The decline of AGTR2 associated with decreased NOS2 mRNA, which consequently affected TNFα, BAX, and CASP3 mRNAs. Apoptosis might be affected by direct effect of AGTR2 on CASP3 during thermal stress. We supposed that NO had a regulatory role during early and late stages of CL. It could be concluded that thermal stress (THI > 68) changed the expression of proliferation and apoptosis genes of CL in Egyptian buffaloes. Finally, the thermal stress in cold or hot seasons has marked impact on CL dynamics.
In this Letter, we provide a novel test of general relativity based on ringdown analysis. The test is performed on agnostic models, where the postmerger signal is fitted with a superposition of damped sinusoids. If at least two modes are detected, one has to compute the ratio of the frequencies and of the damping times and compare them against the predictions of general relativity. By considering ratios, the dependency on the black hole's mass is scaled away. Most notably, we find that the ratios vary very little with the spin, the real part depends mostly on the angular momentum of the mode ℓ and the imaginary part depends mostly on the overtone number n: different combinations create specific mode islands. We provide a qualitative explanation of these islands through a semianalytical argument. We discuss the application of the method to future detectors. Finally, we show that ratios in alternative theories of gravity or between different field content drastically differ from those of general relativity.
Nanoconfined water plays a key role in nanofluidics, electrochemistry, and catalysis, yet its reactivity remains a matter of debate. Prior studies have reported both enhanced and suppressed water self-dissociation relative to the bulk, but without a consistent explanation. Here, using enhanced sampling molecular dynamics with machine-learned potentials trained at first-principles accuracy, we investigate dissociation behavior in water confined within two-dimensional slit pores and nanodroplets, using graphene and hexagonal boron nitride as model materials. We find that reactivity is extremely sensitive to water density, geometry, and surface chemistry, among other factors. Despite this complexity, we show that chemical potential, together with interfacial interactions, governs dissociation trends and explains the variability observed in prior studies. Within this framework, when confined water is compared to the bulk at equivalent chemical potential, corresponding to thermodynamic equilibrium with a bulk reservoir, its reactivity remains essentially unchanged; rather, differences arise when the systems are compared at different chemical potentials or under distinct interfacial conditions. This thermodynamic perspective reconciles previous contradictions and reveals how nanoscale environments can drastically shift water reactivity. Our findings provide molecular-level insight and offer a design lever for modulating water chemistry at the nanoscale.
1. Cholesterol in the central nervous system (CNS) is largely unesterified (>99%) and is predominantly present in the myelin sheath (∼70% of total CNS cholesterol). Damage to the myelin sheath can result in the conversion of cholesterol to cholesterol esters, which occurs in many neurological diseases, including multiple sclerosis. In this study, we measured longitudinal CNS free cholesterol and cholesterol ester levels in a genetic mouse model during postnatal myelination, demyelination, and remyelination using gas chromatography-mass spectrometry with single ion monitoring technique (GC-MS-SIM) and liquid chromatography mass spectrometry (LC-MS). Cholesterol levels in healthy mouse brains increased up to 38 weeks. In contrast, cholesterol in the healthy spinal cord increased during postnatal timepoints, but then remained steady out to 38 weeks. Interestingly, cholesterol esters in the spinal cord were highest at P1 and drastically reduced by P42, while the brain had similar levels during all postnatal time points. During demyelination, both brain and spinal cord cholesterol levels were significantly reduced as compared to healthy mice and failed to return to normal cholesterol levels even during remyelination. Absolute quantification of cholesterol esters during peak demyelination revealed that cholesterol esters comprise 19% of the total cholesterol pool in the brain and 65% in the spinal cord. The lack of recovery in CNS cholesterol levels after demyelination suggests that healthy de novo cholesterol synthesis pathways are disrupted in this model. Absolute quantification of CNS cholesterol is critical for revealing mechanisms of cholesterol regulation during disease and identifying targets for restoring cholesterol to promote myelin repair.
The mitigation of CO2 emissions in petroleum processing requires advanced materials capable of both efficient capture and catalytic conversion to value-added chemicals. In this study, Ni-Sn co-doped graphene (N-Gr@Ni@Sn) was theoretically designed and evaluated as a bifunctional platform for CO2 adsorption and hydrogenation to formic acid (HCOOH). Density functional theory (DFT) calculations, combined with molecular dynamics (MD) simulations, were employed to investigate adsorption geometries, electronic structure modifications, reaction energetics, and stability. Frontier molecular orbital (FMO) and density of states (DOS) analyses revealed a drastically reduced HOMO-LUMO gap (0.189 eV), enhanced electronic conductivity, and the creation of complementary electron-rich and electron-deficient sites around the dopants. Quantum Theory of Atoms in Molecules (QTAIM) and Non-Covalent Interaction (NCI) analyses confirmed a cooperative network of covalent and dispersive interactions stabilizing CO2 at the active sites. The calculated adsorption energy was -225.71 kcal mol-1, with a forward activation barrier of 17.51 kcal mol-1 for desorption and 20.04 kcal mol-1 for CO2-to-HCOOH conversion, indicating favorable thermodynamics and kinetics. MD simulations demonstrated exceptional thermal stability over 1000 ps, with persistent Ni-O(CO2) coordination and Sn-Ni coupling. These findings suggest that Ni-Sn co-doped graphene offers a promising route for integrating CO2 capture with catalytic valorization in petroleum gas treatment, though experimental validation is required to confirm synthesis feasibility, dopant stability, and performance under realistic process conditions.
Pulmonary actinomycosis is a rare cause of granulomatous and suppurative lung disease that often presents with non-specific symptoms leading to misdiagnosis or delayed diagnosis. Complications include sepsis, infective endocarditis and rarely empyema necessitans. In our report, we present a case of a gentleman in his late 50s, with poor oral hygiene and a number of other risk factors who presented with severe weight loss and an expanding chest wall mass. Following a thorough physical examination and biochemical, radiological and microbiological investigations, the patient was diagnosed with an empyema necessitans secondary to pulmonary actinomycosis. The patient was eventually treated successfully with a prolonged course of antibiotics and did not require surgical intervention. Pulmonary actinomycosis has decreased drastically in incidence and the presence of such severe clinical presentations have become even rarer, thanks to the early introduction of antimicrobial agents.
Accurate prediction of the interface residue-residue contacts between interacting proteins is valuable for determining the structure and function of protein complexes. Recent deep learning methods have drastically improved the accuracy of predicting the interface contacts of protein complexes. However, existing methods rely on Multiple Sequence Alignments (MSA) features which pose limitations on prediction accuracy, speed, and computational efficiency. Here, we propose a transformer-powered deep learning method to predict the inter-protein residue-residue contacts using single-sequence and structure-aware protein language models (PLM), called DeepSSInter. Utilizing the intra-protein distance and graph representations and the ESM2 and SaProt PLM, we are able to generate the structure-aware features for the protein receptor, ligand, and complex. These structure-aware features are passed into the ResNet Inception module and the Triangle-aware module to effectively produce the predicted inter-protein contact map. Extensive experiments on both homo- and hetero-dimeric complexes show that our DeepSSInter model significantly improves the performance in both accuracy and speed compared with previous state-of-the-art methods. Integrating predicted contacts significantly improves the docking performance. The DeepSSInter is available at https://github.com/huang-laboratory/DeepSSInter/.
Most organisms carry mobile DNA that enhance their own transmission to subsequent generations, generating conflict with the host genome. The selfish transmission invoked by these selfish genetic elements (SGEs) has promoted a variety of countermeasures by the host genome to reduce their impact. Maternally inherited endosymbionts are common in arthropods and frequently manipulate host reproduction, and transposable elements (TEs) are an exceptionally abundant and diverse group of SGEs. Their activity and abundance vary drastically between even closely related species and can generate evolutionary consequences of both lethal and beneficial effects. Yet, despite their plentitude, many questions remain regarding the potential interactions between different SGEs. This is in part a methodological problem as TEs, for example, often reside in highly repetitive genomic regions, making them difficult to detect. Innovations in genomics have driven renewed interest, particularly with long-read sequencing resolving repetitive regions. We can now begin to define and answer important outstanding questions. For instance, it is unclear how different types of SGEs, including TEs, may interact within host genomes. For example, while different SGEs may compete for host resources (such as availability of molecular machinery), they may also cooperate or even behave parasitically toward each other, as in the case of some TEs. Here we take an "ecology of the genome" approach to examine such interactions that, together with choice examples, may help further our understanding of how interactions between different SGE shape genome evolution.
Efficient and accurate identification of functional genes is critical to biological research, yet traditional single-species approaches are often limited by low efficiency. Previously, we established a novel method for identifying key genes using cross-species protein domain features and machine learning. However, the high multiplicity of gene members associated with specific domains creates a substantial workload for subsequent experimental validation. To address this, this study proposes an enhanced approach that integrates EggNOG-based protein sequence annotation with domain analysis. Unannotated sequences are subsequently analyzed for protein domains, generating a comprehensive "direct gene annotation plus domain" hybrid feature matrix. While the hybrid matrix model yielded comparable predictive accuracy, it significantly enhanced feature resolution: the top 50 predicted features were all known motility-related genes or domains. Furthermore, among the top 100 ranked features, 58 are confirmed to be directly related to motility based on experimental evidence. Although strict genus-level control still yielded 51 confirmed features, excessive taxonomic restriction drastically reduces the number of training genomes, which may paradoxically impair identification efficiency. These results demonstrate that the new method effectively reduces the subsequent experimental workload and enables high-throughput identification of functional genes in a single analysis. With accuracy and efficiency far exceeding those of existing single-species identification methods, it provides a highly efficient solution for mining key genes underlying other complex bacterial phenotypes.