中国人民解放军军事科学院军事医学研究院

Respiratory vaccinology has long prioritized the discovery of increasingly potent immune adjuvants, yet this strategy has failed to resolve the core limitation of mucosal vaccination: the lack of spatiotemporal control over innate immune activation. NanoCF501, a precision-engineered inhalable STING nanoadjuvant recently reported by Liu et al., signals a paradigm shift. Its transformative value lies not in molecular potency, but in the rational integration of particle architecture, size tuning (∼18 nm for optimal mucus penetration), polymer chemistry, and pharmacokinetic compartmentalization to program immune outcomes. By achieving coordinated mucosal–systemic immunity at one-twentieth the systemic dose while eliminating systemic exposure, NanoCF501 demonstrates that mucosal vaccine design is fundamentally a challenge of precision engineering rather than empirical adjuvant discovery. We distill four design principles from this work and contextualize them within a broader emerging framework of sequential biological barrier engineering, from aerosol stability to intracellular delivery.

Accurate prediction of organ-specific toxicity with mechanistic interpretability remains a central challenge in chemical safety assessment and translational toxicology. Although animal-based assays provide biologically relevant information, they are low throughput and increasingly constrained by ethical and regulatory considerations. In parallel, data-driven computational approaches-particularly deep learning-have achieved strong predictive performance but often lack transparency, limiting their utility for mechanistic interpretation and regulatory decision-making. Here, we present ToxiGuard, a mechanistically informed deep learning framework that embeds organ-specific Adverse Outcome Pathway (AOP) structures directly into the model architecture. By explicitly encoding Molecular Initiating Event-Key Event-Adverse Outcome (MIE-KE-AO) connectivity, ToxiGuard constrains representation learning to biologically plausible causal pathways rather than relying solely on post hoc explanation. The framework integrates molecular descriptors, functional-class fingerprints, and curated AOP networks to enable complementary interpretability at both chemical and biological levels. Across four organ toxicity endpoints-hepatotoxicity, cardiotoxicity, nephrotoxicity, and respiratory toxicity-ToxiGuard demonstrates robust and consistent predictive performance, outperforming traditional machine learning models and AOP-agnostic neural network baselines. SHAP-based analyses reveal concentrated contributions from specific physicochemical properties, molecular substructures, and mechanistic AOP components, including CAR/PXR-associated pathways in hepatotoxicity and hERG-related mechanisms in cardiotoxicity. These findings are consistent with established toxicological knowledge and support the biological plausibility of model predictions. Overall, this study demonstrates that embedding mechanistic toxicological structure into deep learning architectures can enhance both predictive reliability and interpretability. ToxiGuard provides a transparent, AOP-anchored computational approach that complements existing experimental and in silico methods for early-stage organ toxicity assessment and chemical safety evaluation.

Advanced Oxidation Processes (AOPs) based on peroxymonosulfate activation hold promise for degrading organic pollutants, nevertheless challenges remain in enhancing free radical utilization and achieving deep water purification. Herein, we designed a Co-doped W18O49 (CWO) catalyst to address these challenges by boosting the generation of SO4 center dot- while reducing their interaction distance with pollutants. Under light excitation, the incorporation of Co sites enhanced photoelectron transfer, promoted the desorption of *SO4 intermediates, accordingly increase the generation of SO4 center dot-. Simultaneously, the distribution of asymmetric charges enhances the local dipole moment of the catalyst, shortening the interaction distance between pollutants and SO4 center dot- through dipoledipole interactions. Based on this, the k-value of carbamazepine degradation in the CWO system was 20-fold compared to the WO system. Furthermore, in the CWO@PVA-AL hydrogel system, the COD removal rate for mixed wastewater reached up to 92%. The results of Electrical Energy per Order (EE/O) metric and Life Cycle Assessment (LCA) of the system driven by sunlight demonstrated the practical applicability of this highly efficient free radical system.

Designing heuristics for combinatorial optimization problems (COPs) is a fundamental yet challenging task that traditionally requires extensive domain expertise. Recently, Large Language Model (LLM)-based Automated Heuristic Design (AHD) has shown promise in autonomously generating heuristic components with minimal human intervention. However, most existing LLM-based AHD methods enforce fixed algorithmic templates to ensure executability, which confines the search to component-level tuning and limits system-level algorithmic expressiveness. To enable open-ended solver synthesis beyond rigid templates, we propose Automated Algorithm Design via Evolutionary Program Trees (A2DEPT), which treats LLMs as system-level algorithm architects. A2DEPT explores the vast program space via a tree-structured evolutionary search with hybrid selection and hierarchical operators, enabling iterative refinement of complete algorithms. To make open-ended generation practical, we enforce executability with a lightweight program-maintenance loop that performs feedback-driven repair. In experiments, A2DEPT consistently outperforms representative LLM-based baselines on both standard and highly constrained benchmarks. On the standard benchmarks, it reduces the mean normalized optimality gap by 9.8

Selective aerobic oxidation of thioethers to sulfoxides over efficient visible-light photocatalysts is an essential yet highly challenging transformation in catalytic oxidation chemistry. Herein, we introduce an asymmetric perylene diimide-based it-conjugated polymer (BCPDI) with intrinsic polarization that enhances charge transport and tunes the local microenvironment to efficiently regulate reactive oxygen species(ROS) generation for efficient thioether oxidation. DFT calculations show that symmetry breaking separates frontier orbitals and creates an internal electrostatic potential gradient, thereby accelerating charge carrier transport and driving O2 reduction to generate superoxide radicals (center dot O2-), whereas symmetric analogs produce only singlet oxygen (1O2) via energy transfer. Driven by the synergistic effect of electron and energy transfer pathways, BCPDI achieved over 99% conversion and selectivity for thioether oxidation under ambient conditions, outperforming most recently reported materials. In addition, BCPDI also exhibits a superior catalytic activity for thioether oxidation (e.g., aromatic thioethers, aliphatic sulfide, and Mustard Gas sulfide). This study provides a new approach for the design of catalysts for photocatalytic aerobic oxidation of thioethers and deepens the understanding of thioether activation mechanisms mediated by photogenerated charge carriers and reactive oxygen species.