Hemifacial microsomia (HFM) presents with complex hard and soft tissue deficiencies that challenge reconstruction, especially when mandibular segments are absent. An early childhood patient with right-sided hemi-mandibular hypoplasia (not consistent with the current Pruzansky-Kaban Classification but close to type III) and a rudimentary condyle underwent mandibular reconstruction using a costochondral graft extending from the body to the rudimentary condyle restoring the entire ramal portion. Postoperatively, the patient achieved improved facial symmetry, enhanced mandibular movements and better masticatory function. Patient follow-up was kept at 1st, 3rd, 6th and 12th months. The graft showed good integration, and preservation of the native condyle facilitated growth and functional restoration in comparison to distraction osteogenesis as an option for corrective surgery. Costochondral graft reconstruction provides a biologically favourable solution in growing HFM patients, though future corrective surgeries may be required to refine facial symmetry and occlusal balance as growth continues.

Ultrasound imaging, with its advantages of portability, real-time feedback, and non-invasiveness, has become an indispensable modality in computer-aided diagnosis. However, inherent challenges in ultrasound images, such as speckle noise, low contrast, and blurry boundaries, significantly hinder segmentation accuracy. Existing methods attempt to alleviate these issues by deepening network architectures, enlarging convolutional kernels, or introducing self-attention mechanisms to enhance feature representation. Nevertheless, these strategies often compromise computational efficiency, making it difficult to achieve a desirable balance between segmentation accuracy and inference speed, thereby limiting their applicability in resource-constrained clinical scenarios such as portable ultrasound devices. To address these challenges, this paper proposes a lightweight and efficient segmentation network, termed LS2Net. In the encoder, to mitigate the impact of speckle noise, LS2Net incorporates wavelet convolution to transform spatial features into the frequency domain for targeted processing, which not only suppresses noise but also enlarges the receptive field and captures global contextual information. Meanwhile, pixel difference convolution is employed to capture fine-grained details under low-contrast conditions, enhancing the perception of local textures. Furthermore, we design a Residual Refinement Skip Connection (RRSC) module, which leverages the discrepancy between down-sampled and up-sampled features to preserve informative components while filtering redundant information, thereby facilitating more effective feature reconstruction. In the decoder, a Multi-Receptive Field Coarse-to-Fine Module (MRCFM) is introduced to further integrate multi-scale contextual information through hybrid receptive fields, enabling more precise delineation of object boundaries. LS2Net demonstrates superior computational efficiency and state-of-the-art (SOTA) performance across five ultrasound datasets, achieving a favorable balance between segmentation accuracy and inference speed. In addition, it delivers competitive performance on other imaging modalities, highlighting its versatility. Moreover, LS2Net exhibits strong generalization capability in cross-dataset evaluations, validating its robustness and scalability. Overall, LS2Net provides a practical and deployable solution for real-time clinical applications, particularly suitable for portable or resource-limited devices. The source code is publicly available at: https://github.com/CYYJL/LS2Net.

Early flame detection via Unmanned Aerial Vehicles (UAVs) is crucial for wildfire prevention. However, extreme small scales, non-rigid morphological distortions, and severe background interference in aerial imagery, coupled with limited edge-computing power, pose significant challenges.

Janus droplets enable advanced electrokinetic transport due to their intrinsic interfacial asymmetry, yet most existing systems rely on permanently engineered surface heterogeneity that limits dynamic and reversible control over electrohydrodynamic behavior. Here, we propose a zwitterionically functionalized Janus droplet whose interfacial charge can be programmably regulated through solution pH, enabling chemically tunable induced-charge electrophoretic (ICEP) dynamics under an applied electric field. The pH-responsive zwitterionic coating modulates the balance between electroosmotic slip and electrophoretic motion, leading to reversible transitions between vortex-dominated momentum-dissipative states and streamlined high-mobility transport regimes. Systematic numerical simulations reveal that the extent of zwitterionic coverage, grafting density, droplet size, and electric-field polarity collectively govern ICEP vortex formation and propulsion characteristics. Under acidic conditions, the positively charged interface promotes strong counter-rotating ICEP vortices that enhance viscous dissipation and reduce droplet velocity, whereas under basic conditions the reversal of interfacial charge aligns electroosmotic and electrophoretic transport, suppressing vortices and enhancing propulsion. Reversal of the electric-field polarity enables dynamic switching of vortex formation across pH conditions, providing an additional mechanism for externally controlled transport modulation. To rationalize these behaviors, we further develop a scaling-based analytical framework that classifies the electrohydrodynamic response into wall-dominated, vortex-dominated, and slip-dominated regimes, yielding predictive transition criteria that agree well with numerical observations. The present study establishes charge-regulated zwitterionic functionalization as a viable strategy for reconfigurable electrokinetic transport and adaptive flow manipulation in microfluidic systems, with potential applications in controllable mixing, separation, and programmable droplet transport.

Beyond its classical role as the cellular energy currency, adenosine triphosphate (ATP) can directly modulate protein stability, solubility and self-assembly through weak nonspecific interactions. Here, we summarise the current experimental and theoretical understanding of noncanonical ATP-protein interactions and their implications for folding, aggregation and phase stability. Recent studies show that ATP interacts primarily with positively charged residues such as lysine (Lys) and arginine (Arg) via electrostatic and hydrogen-bonding interactions, with additional cation-π or π-π contacts in some systems. These interactions modulate protein folding, conformational stability and aggregation by shifting the balance between folded and unfolded states. For intrinsically disordered and amyloidogenic proteins, ATP acts as a concentration-dependent regulator that can promote, suppress or remodel phase-separated condensates and fibrillar assemblies depending on protein sequence, ATP concentration and solution conditions. In natively folded proteins, ATP influences protein-protein interactions and phase behaviour, often enhancing colloidal stability and suppressing aggregation. Complexation with Mg[2+] further modulates these effects by altering phosphate charge accessibility and, in some systems, changing the balance between binding, bridging and solubilisation. Together, these observations establish ATP as a multivalent small-molecule modulator of protein behaviour.

To address strong matrix interference and weak signals in ppb-level Raman detection of mycotoxins in wheat, this study developed a synergistic strategy combining the microfluidic "hard processing" with the feature selection "soft processing". A polydimethylsiloxane microfluidic chip was designed to separate and enrich toxins based on solubility differences. Combined with the Monte Carlo sampling - weighted bootstrap sampling - least absolute shrinkage and selection operator feature selection algorithm, specific spectral information was extracted to build a partial least squares regression model. The results demonstrate that this strategy effectively eliminates matrix interference from starch and proteins, and the estimated limits of detection for all four mycotoxins are below the Chinese national regulatory limits (GB 2761-2017), with test set coefficients of determination all exceeding 0.92. Feature dimensionality was reduced from thousands to under one hundred, focusing on mycotoxin core functional group vibrational peaks. This approach balances accuracy, stability, and efficiency. With further validation across multiple scenarios, it holds promise for development into a standardized detection solution for such trace contaminants.

Whole Slide Image (WSI) analysis, with its ability to reveal detailed tissue structures in magnified views, plays a crucial role in cancer diagnosis and prognosis. Due to their giga-sized nature, WSIs require substantial storage and computational resources for processing and training predictive models. With the rapid increase in WSIs used in clinics and hospitals, there is a growing need for a continual learning system that can efficiently process and adapt existing models to new tasks without retraining or fine-tuning on previous tasks. Such a system must balance resource efficiency with high performance. In this study, we introduce COSFormer, a Transformer-based continual learning framework tailored for multi-task WSI analysis. COSFormer is designed to learn sequentially from new tasks wile avoiding the need to revisit full historical datasets. We evaluate COSFormer on a sequence of seven WSI datasets covering seven organs and six WSI-related tasks under both class-incremental and task-incremental settings. The results demonstrate COSFormer's superior generalizability and effectiveness compared to existing continual learning frameworks, establishing it as a robust solution for continual WSI analysis in clinical applications. The code is released at https://github.com/QuIIL/COSFormer.

The effectiveness factor (EF) concept is extended for non-catalytic pyrolysis reactions involving no gaseous reactant. A fully transient, non-isothermal model is formulated for a single biomass pellet by incorporating structural evolution, porosity variation, and temperature-dependent properties. The study presented a theoretical analysis of EF for different values of Lewis number (L e), heat of reaction (G factor), Arrhenius number (Ē), Thiele modulus (ϕ), and heating rate (γ). The model was solved using two approaches (i) a numerical solution and (ii) an approximate solution based on the Quantized Method (QM). Good agreement was observed between the numerical and QM solutions for pellet conversion within the investigated thermicity factor range of -0.25 ≤ β ≤ 0.225, while the effectiveness factors predicted by the two approaches differed by a maximum of 7.14%. Furthermore, analytical asymptotic limits show that EF approaches unity for homogeneous behavior, while it drops to as low as 0.005 (200-fold reduction in reaction rate) under shrinking-core conditions. For practical applications in fixed-bed gasifiers, a pellet diameter of 4 mm is recommended under the investigated conditions (L e = 0.25, G = -6), providing a favorable balance between heat and hydrodynamic constraints. The developed EF correlations enable rapid estimation of effective pyrolysis rates without detailed multiscale modeling, offering a computationally efficient tool for reactor design and optimization in biomass gasification systems.

Synthetic biomolecular condensates offer a route to engineer compartmentalized microenvironments with tunable physicochemical properties, yet design principles for controlling their internal organization remain limited. Here, we report a minimal two-component system based on thermoresponsive intrinsically disordered protein polymers with lower critical solution temperature behavior (LCST-IDPPs), in which electrostatic topology programs liquid-liquid phase separation (LLPS). Incorporating charged residues into LCST-IDPPs suppresses phase separation under physiological conditions, whereas mixing oppositely charged, individually non-coacervating IDPPs restores LLPS as an emergent, composition-dependent process driven by multivalent intermolecular charge compensation. We compare this two-component system with a covalently linked diblock containing the same charged chains. This change in chain connectivity alters the coupling between electrostatic pairing, counterion redistribution, and LCST-driven dehydration. As a result, the balance between inter- and intrachain ionic pairing encodes the residual charge and micropolarity of the dense phase. This topology-dependent microenvironment controls condensate miscibility and drives the formation of either homogeneous or internally demixed multiphase assemblies. The condensate interior also shifts the apparent pKa of ionizable residues, indicating that phase separation modifies local acid-base equilibria and alters the effective side-chain charge. Together, these findings show how electrostatic topology influences LLPS, the dense-phase microenvironment, and mesoscale organization in IDPP condensates.

In this study, a dimethyl sulfoxide (DMSO)-incorporated low-dimensional butylammonium lead iodide hybrid crystal was synthesized at room temperature using a mixed dimethylformamide/dimethyl sulfoxide (DMF/DMSO) solution-growth method. Single-crystal X-ray diffraction (SC-XRD) analysis revealed that the material does not form a three-dimensional corner-sharing perovskite framework, but instead consists of a low-dimensional BA-Pb-I hybrid structure containing a one-dimensional Pb-I ribbon-like inorganic motif. The crystallographic formula was determined to be (C4H12N)2Pb3I8·2(C2H6OS), confirming a charge-balanced composition with DMSO incorporated as a neutral molecule within the crystal lattice. The structural and optical properties were investigated using X-ray diffraction (XRD), electron microscopy, ultraviolet-visible (UV-Vis) absorption spectroscopy, temperature-dependent photoluminescence (PL), and time-resolved PL decay measurements. The crystal exhibited two low-temperature emission peaks near 487 and 512 nm, comparable to excitonic emission features reported for BA2PbI4-based low-dimensional systems. Temperature-dependent PL analysis showed reduced exciton-phonon coupling and weaker thermally induced linewidth broadening compared with reported BA2PbI4 surface emission. High-temperature XRD and differential scanning calorimetry (DSC) measurements further supported stable structural and thermal behavior under the tested conditions. These results provide insight into the room-temperature mixed-solvent crystallization, structural characteristics, and excitonic optical behavior of DMSO-incorporated low-dimensional BA-Pb-I hybrid crystals.

We present a novel diversity-focused haplotype map (HapMap) that characterizes over 64.5 million maize (Zea mays ssp. mays) single nucleotide polymorphisms (SNPs) genotyped across 818 individuals from diverse backgrounds. This HapMap aims to balance the variation obtained from domesticated landraces and inbred lines, outgroup Zea spp. and more distant Tripsacum spp. in order to minimize ascertainment bias for diversity studies. Included individuals derive from public data from various experimental setups and coverages, which is challenging for standard SNP callers to accommodate. We provide evidence of coverage biases associated with standard callers that influence resulting variation and introduce a novel approach called Unified Multi-Caller Ensemble (UME), which enhances variant calling accuracy in low-coverage and mixed-coverage genomic datasets. UME corrects for coverage bias resulting from inter-sample coverage heterogeneity by leveraging evidence from variant callers with orthogonal strategies, re-calibrating the error probabilities across callers to minimize the impact of error biases inherent to a given caller. It outperforms individual strategies and excels in de novo variant calling, taking advantage of instances of higher depth reads, even in low coverage individuals, while preserving biologically informative variant relationships across coverage levels. An important feature of UME is the independence from population allele frequencies in the discovery panel, thus avoiding ascertainment bias resulting from unbalanced input genetic diversity. Discovered variants are less affected by ascertainment bias because no population filtering is used, and the full diversity of SNPs is retained in the final variant call set to maximize the utility of the dataset for production calling newly sequenced samples. We present a strategy for filtering the recalibrated error profiles that relies on maximizing demographic signals to retain genetic relationships within the population while reducing sequencing error. After the variant discovery phase, we employ the UME production stage, which enriches genotype calling across all coverage levels, benefiting low-coverage samples. Error introduced in this process is removed through subsequent filtering. Using this approach, we generated a coverage bias-controlled maize HapMap database, providing a comprehensive representation of maize accessions and emphasizing landrace diversity. This diverse panel of domesticated maize and outgroups from across the Americas enables accurate genotyping in low-coverage samples while offering crucial context for interpreting diversity, particularly for natural diversity and paleogenomic analyses.