Cell segmentation in imaging-based spatial transcriptomics (ST) relies on stained image segmentation or transcript clustering, but current methods are limited by dependence on manual annotations, sensitivity to staining quality and cell density, and lack of alignment with true transcriptomic profiles. We introduce RedeFISH, a reinforcement learning-based framework that formulates cell segmentation as a sequential strategic decision-making process. Through iterative optimization of the transcript assignment policy, RedeFISH aligns segmented cell expression profiles with single-cell expressions to achieve optimal segmentation. This approach enables staining-free, transcriptome-informed delineation of individual cells, thereby overcoming key limitations of existing methods. Benchmarking across multiple ST platforms shows that RedeFISH improves cosine similarity of segmented cell expression profiles by 9.7% on average and reduces root mean squared error (RMSE) by 13.9% compared with state-of-the-art methods. It also increases agreement between segmented and ground-truth cell regions by 19%, demonstrating improved accuracy and robustness. RedeFISH further imputes whole-transcriptome profiles from sparse ST measurements via spatially guided transfer from scRNA-seq data. The whole-transcriptome coverage enables unbiased and in situ prioritization of the critical regulators of spatial niches, e.g., mouse intestinal stem cell niches, and unlocks the possibility to delineate, with only one sample, the panoramic molecular and cellular changes along the whole developmental process from initiation to invasion for human cancer.

Aircraft serve as the intelligent extension of human capabilities into 3-dimensional space. Next-generation aircraft will transcend the limitations of traditional fixed aerodynamic configurations, evolving into embodied intelligent agents capable of sensing their environment, making autonomous decisions, acting collaboratively, and self-evolving—ultimately achieving Fly-by-Feel. Flexible sensing skins, with their unique advantages (ultrathin, lightweight, large-area conformal attachment to the aircraft surface, and adaptive regulation), are emerging as the key enabling technologies. Multi-aerodynamic parameters are acquired through in situ sensing, and artificial intelligence (AI)-enhanced decision-making is implemented via edge-deployed chips. Based on this sensing–algorithm collaborative bridge, closed-loop actuation is further realized to support functional implementation. This paper reviews the core composition of the “sensing–decision–regulation” closed-loop system, including aerodynamic disturbance-free in situ sensing and AI-enhanced super-resolution field perception, and demonstrates its applications in 3 key scenarios: anti-icing/deicing, electromagnetic stealth, and lift enhancement/drag reduction. Finally, it highlights the future prospects of programmable intelligent flexible skins, promoting the aircraft’s paradigm shift toward intelligence, efficiency, and safety.

Chiral structures have substantial potential for spin information utilization because of their distinctive spatial symmetry-breaking features and the consequent strong spin–charge correlation. One of the limitations in the development of chiral spintronics for both theoretical and practical applications is the limited range of chiral-related signal transmission, which is only tens of nanometers and is usually determined by the conductivity of the chiral materials and device applicability. In addition, the role of circularly polarized light, an important chiral parameter, in chiral signal transmission has been typically neglected. In this study, targeted chiral induction was conducted using the semiconductor polymer PCDTPT ([4-(4,4-dihexadecyl-4H-cyclopenta[1,2-b:5,4-b′]dithiophen-2-yl)-alt-[1,2,5]thiad-iazolo [3,4-c] pyridine]), which was further used to fabricate a field-effect transistor device. In a dark environment, the chiral-related signal could be transmitted up to 10 μm with an external magnetic field. Through the synergistic effects of spin-selective transition and chiral spin filtering under illumination, the chiral signal expression was fundamentally switched to another form, and the transmission distance was notably increased to the millimeter scale. In addition, owing to the flexible external field tunability (gate voltage, temperature, and polarized light) in multiple scenarios, the chiral signals reflected detailed dynamic changes. The achievement of long-distance chiral signal transmission and circularly polarized light-induced signal conversion and extension is expected to promote the development of chiral spintronics for both theoretical and practical applications.

The structural design of bone scaffolds determines the local fluid mechanical microenvironment; however, how such cues program stem cell metabolism to drive osteogenesis remains unclear. In this study, Voronoi-based trabecular-like scaffolds with tunable porosity were engineered to modulate fluid shear stress (FSS) while preserving a consistent topology. Computational fluid dynamics analyses confirmed that architectures with lower porosity generated higher FSS, enabling controlled investigation of mechano-metabolic coupling. Under dynamic culture conditions, bone marrow mesenchymal stem cells (BMSCs) cultured on high-FSS scaffolds exhibited enhanced osteogenic differentiation in vitro and promoted bone regeneration in vivo. Integrated transcriptomic, proteomic, and metabolomic analyses identified caveolin-1 (CAV1) as a prominent FSS-responsive membrane regulator. Mechanistically, CAV1 enhanced phosphatidylinositol 3-kinase (PI3K)–AKT signaling, stabilized hypoxia-inducible factor-1α (HIF-1α), and induced a glycolytic shift that supports the energetic and biosynthetic demands of osteogenesis. Pharmacological inhibition of PI3K, HIF-1α, or glycolysis abolished FSS-driven osteogenic responses, validating a CAV1-centered mechano-metabolic axis. These findings establish a direct link between scaffold microarchitecture and metabolic regulation of osteogenesis and provide design principles for mechanically instructive bone repair materials.

Developing advanced materials with simultaneously excellent wave transparency and efficient thermal insulation is critical for hypersonic vehicles. While rare earth disilicates (RE₂Si₂O₇) are promising candidates, their vast chemical space and complex polymorphism hinder precise property modulation. Herein, we establish an integrated high-throughput experimental and machine learning strategy to systematically investigate the composition–structure–property relationship of high-entropy RE₂Si₂O₇. The results demonstrate that the average RE³⁺ ionic radius determines the phase boundary. Notably, Sc incorporation jointly reduces both the dielectric constant and thermal conductivity. Specifically, the small size and strong electron localization of Sc minimize the polarizability, while its severe size and mass mismatch with other RE elements intensify phonon scattering. The model’s generalization is further validated by designing a series of high-entropy RE₂Si₂O₇ containing 5 to 9 distinct RE elements. Ultimately, the (Ho_1/5Tm_1/5Yb_1/5Lu_1/5Sc_1/5)₂Si₂O₇ high-entropy ceramic achieves a low dielectric constant (ε = 5.4) and a low thermal conductivity (κ = 1.3 W·m⁻¹·K⁻¹). This data-driven strategy provides a new pathway for the rational design of advanced high-entropy wave-transparent materials for extreme environments.

Wearable electronics serve as critical tools for human health monitoring and equipment operation assessment. A fundamental prerequisite for their stable performance is the ability to attach flexibly and reliably to the surfaces of human bodies or equipment. However, existing adhesion methods for wearable devices face substantial challenges in simultaneously achieving high-strength reversible bonding and environmental adaptability across diverse conditions, hindering the practical applications in complex and variable environments. Here, inspired by the excellent adhesion behaviors of geckos and octopuses, we designed a dual bioinspired adhesive microstructure consisting of an annular stalk, a microdome, and an annular tip. The large-area precise formation of this complex 3-dimensional microstructure is achieved via simple imprinting and photolithography. The fabricated adhesive demonstrated reversible and superior normal adhesion in complex environments including vacuum (120 kPa), dry (201 kPa), moist (173 kPa), and underwater (165 kPa). In situ observations of the contact splitting reveal that its extraordinary performance derives from the unique crack reentry phenomenon across various environments, which effectively inhibits crack propagation and improves the adhesion forces. Simulations further clarify the causes of crack nucleation and the underlying adhesion mechanisms. Demonstrations of the adhesive in skin-attachable electronics across multiple environments highlight its potential for applications in wearable electronics operating under diverse and complex conditions, promoting rapid development of wearable flexible electronics.

Malignant pleural mesothelioma (MPM) is a rare and highly aggressive malignancy arising from the pleural lining, characterized by a dismal prognosis and limited therapeutic options. Mesothelin (MSLN)-directed chimeric antigen receptor (CAR)-armored T (CAR-T) cell therapies have shown encouraging preliminary outcomes; however, autologous manufacturing approaches remain constrained by logistical complexity and prolonged production timelines, which are suboptimal for patients with rapidly advancing disease. Here, we describe the development of human allogeneic interleukin-15-augmented, MSLN-specific, CAR-armored invariant natural killer T (^Allo15MCAR-NKT) cells. These cells are generated through genetic modification of hematopoietic stem and progenitor cells, followed by a clinically guided CAR-NKT cell differentiation, maturation, and expansion process. This approach supports scalable production with high cellular yield, purity, and translational feasibility. Functionally, ^Allo15MCAR-NKT cells exhibit robust antitumor efficacy in vitro and demonstrate robust therapeutic activity across multiple in vivo MPM xenograft models, including subcutaneous and lung metastasis models. In addition, they actively modulate the tumor microenvironment by targeting CD1d⁺ tumor-associated macrophages. Phenotypic analysis reveals a rejuvenated cellular profile, marked by low expression of exhaustion-associated and inhibitory receptors, including PD-1, TIM-3, LAG-3, CTLA-4, and TIGIT, consistent with sustained functional capacity. Importantly, these cells display a favorable safety profile, with minimal evidence of graft-versus-host disease, cytokine release syndrome, brain infiltration or neurotoxicity, and no detectable off-tumor effects. Collectively, these findings support the development of a clinically translatable, off-the-shelf CAR-NKT cell therapy for the treatment of MPM.

Major histocompatibility complex class II (MHC-II) neoantigens play a critical role in immunotherapy, either as direct effectors or through their influence on CD8⁺ T cells. However, only a small fraction of tumor DNA mutations qualify as functional neoantigens, and current prediction tools suffer from limited accuracy, making the in vivo design of highly immunogenic neoantigens challenging. Here, we present EpiMII, a structure-aware graph neural network model for de novo design of MHC-II epitopes. By integrating 3-dimensional structural information through an inverse folding strategy, EpiMII generates mimotopes that preserve T cell specificity while enhancing MHC-II binding affinity and immunogenicity. The model was trained on a first-of-its-kind curated dataset of 142,934 homology-modeled MHC-II epitope structures, overcoming the scarcity of experimentally resolved MHC-II–peptide complexes. Benchmarking analyses illustrate that EpiMII achieves a sequence recovery rate of 66.7% on a held-out test set and 79.0% on crystallized epitopes, markedly outperforming existing tools like ProteinMPNN. In a hepatocellular carcinoma case study, all 5 EpiMII-designed epitopes could markedly activate CD4⁺ T cells in vitro and induce the secretion of interferon-γ and tumor necrosis factor-α. Notably, treatment with one epitope (P4) substantially reduced tumor volume in mice in vivo, comparable to a known positive control. These findings establish EpiMII as a powerful tool for structure-guided neoantigen discovery, with broad implications for cancer vaccine development and personalized immunotherapy.

The most common form of hair loss, androgenetic alopecia (AGA), is characterized by progressive follicular shrinkage. Current research efforts are focused on developing novel therapeutic techniques that exhibit better efficacy and safety profiles. In response to this challenge, biomedical technology has emerged as a multidisciplinary field and offered marked advantages in overcoming skin barriers, enabling controlled drug release, and modulating the perifollicular microenvironment. This review provides a comprehensive overview of advanced biomedical technologies for AGA treatment. We first outline the pathogenesis and introduce some conventional therapies. We then delve into the advantages and practical applications of emerging biomedical technologies, including novel pharmacological formulations, nanotechnology, stem cell-based approaches, and microneedle platforms. Subsequently, our focus shifts to the analysis of current therapeutic strategies in addressing core pathogenic factors through the latest and representative case studies. The review concludes by highlighting major challenges and future directions of these biomedical technology-based approaches, aiming to provide insights for the rational design of next-generation AGA therapies.

The efficacy of prostate-specific membrane antigen (PSMA)-targeted radioligand therapy in metastatic castration-resistant prostate cancer is limited by tumor heterogeneity and intrinsic radioresistance. This study investigates 2 natural flavonoids, genistein and baicalein, as potential radiosensitizers to improve the therapeutic effectiveness of ¹⁷⁷Lu-PSMA617 therapy. In vitro, combined treatment with low-toxicity concentrations (5 to 15 μM), based on cellular median inhibitory concentration (IC₅₀) of either flavonoid, dose-dependently enhanced the cytotoxicity of ¹⁷⁷Lu-PSMA617 against the human prostate cancer cell line LNCaP, reducing its IC₅₀ to 14.2% to 45.1% of that with monotherapy. All combination regimens demonstrated synergistic interaction, with combination indices below 0.82. Mechanistically, the combinations (particularly at 10 μM) induced a pro-oxidant shift (increasing reactive oxygen species by 78% to 144%), amplified DNA damage (γ-H2AX increased by 85% to 115%), and promoted apoptosis (caspase-3 activity increased by 325% to 389%) while suppressing prosurvival p-AKT. Transcriptomic profiling further revealed that both flavonoids induced a pro-apoptotic gene signature and markedly modulated multiple cancer-related pathways, particularly inhibiting key DNA double-strand break repair pathways such as nonhomologous end-joining. In vivo, combination therapies profoundly inhibited tumor growth (50.2% to 78.1% reduction versus monotherapy) and extended survival, with no overt systemic toxicity observed in this preliminary assessment. This work establishes genistein and baicalein as effective, multi-targeted radiosensitizers that enhance ¹⁷⁷Lu-PSMA617 efficacy by cooperatively amplifying DNA damage, inhibiting repair mechanisms, and activating the apoptotic cascade. Their natural origin and established safety profile underscore their translational potential. Thus, genistein and baicalein represent promising novel adjuvants for prostate cancer radioligand therapy.