Data-driven generative design of lattice metamaterials often demands excessively large training datasets to capture complex three-dimensional geometries. This work introduces an exploratory hierarchical workflow that circumvents this limitation through a compact, cubic symmetry-constrained topological representation, enabling efficient design space exploration with minimal data and a lightweight machine learning surrogate. Using 213 additively manufactured sub-millimetre AlSi10Mg specimens, the approach identifies two novel lattice topologies: fractal face cubic and cross face cubic. These metallic architectures achieve a significant increase of up to 30% higher yield strength compared to simple cubic lattices, a widely used high-strength lattice benchmark (stretch-dominated), across multiple relative densities and a broad range of strut slenderness ratios, covering distinct deformation regimes. Notably, the emergent fractal face cubic topology reveals an unprescribed transition toward thin-plate-like architectures, where strut fusion promotes uniform load distribution and maximizes structural efficiency. This small-but-efficient-data framework provides a scalable path for discovering non-intuitive, high-performance architected materials within symmetry-constrained design spaces, with direct relevance to thin-walled structural applications.

Antimicrobial peptides (AMPs) are emerging as potential antibacterial drugs in the face of rapidly increasing bacterial resistance to conventional antibiotics. These peptides target the microbial membrane, and numerous models are used to describe their mechanism of action, causing cell content release and membrane disruption leading to cell death. The interaction between AMPs and target membranes is critical to their specificity and activity. However, a precise understanding of the relationship between AMP structure and their cytolytic function in a range of organisms remains elusive. The challenges in the field reside in the complex nature of AMP interactions with cell membranes, the mechanism of which varies considerably between different AMP classes. Biophysical techniques are widely used to study the influence of peptide structures and membrane properties on AMP cytolytic activity in model membrane systems and, recently, in live cells. In this review, we discuss high-resolution techniques that are employed to study the mechanism of membrane disruption and provide structural and functional insights into AMPs. Our aim is to provide a compendium of experimental and theoretical modeling approaches that can be used to characterize the folding and interactions of AMPs and synergistically advance the development of a new generation of effective antimicrobial therapeutics.

Antimicrobial peptides (AMPs) offer promising alternatives to conventional antibiotics, yet most predictive models fail to account for chemical modifications that influence real-world efficacy. Among these, C-terminal amidation is a widely adopted and effective strategy that improves structural stability, membrane interaction, and protease resistance. In this study, we established an integrated framework for the design and prediction of C-terminal amidated AMPs targeting Escherichia coli. Our approach combined a design-oriented model based on an interpretable Explainable Boosting Machine (EBM), which extracts actionable sequence-level design rules, together with a reliable deployment model, built on a fine-tuned ESM2 deep learning architecture. The resulting tool, CAmidPred, enables both predictive classification and amino acid pattern analysis with outputs examined in relation to published alanine-scanning experiments. Using these models, we identified a pardaxin variant with improved activity against E. coli, demonstrating the practical utility of the dual-model framework in targeted AMP design.

Protic organic salts have great potential to be used as phase change materials for thermal energy storage. However, tuning their melting temperatures and maximising their energy storage density (enthalpy of fusion) is a great challenge. The structures of the cation and anion play a crucial role in determining the thermal properties of protic organic salts. In this study, linear and non-linear machine learning models are used to predict the melting temperature (Tm) and enthalpy of fusion (Delta Hf) of 182 possible protic salts using thermal properties (Tm and Delta Hf) of 69 protic salts for training models. An additional feature of this study was the investigation of the prediction accuracy of models for salts with solid-solid phase transitions. It was found that the presence of solid-solid transition/s greatly impacted the Delta Hf predictions. The best linear models for Delta Hf were obtained for salts having no solid-solid transitions (R2 of 0.82, standard error of estimation (SEE) of 4 kJ mol-1). Tm predictions remained unaffected by the presence of solid-solid transitions. The best linear model for Tm prediction achieved R2 of 0.63, and SEE of 28 degrees C. The non-linear models showed marginally lower performance compared to linear models. Experimental cross-validation demonstrated the acceptable predictive ability of linear models for both Tm and Delta Hf. This study opens new avenues for exploring the molecular origins of PCM properties and advancing the development of efficient energy storage materials.

Machine learning (ML) has emerged as a transformative tool for the design and optimization of functional materials, offering significant potential to accelerate the discovery and improve performance. In the field of surface coatings, although still in its early stages, ML is increasingly being applied to create novel coating materials with enhanced properties such as adhesion, hardness, durability, and corrosion inhibition. By using data-driven approaches, researchers can optimize formulations and processing conditions more efficiently than traditional trial-and-error methods, paving the way for innovation in advanced coatings that meet specific requirements. This review paper explores the current applications of ML in surface coating research, emphasizing successful case studies that demonstrate its effectiveness in tailoring and enhancing coating properties. The paper also identifies key opportunities and challenges for further integrating ML into coating design workflows.

The development of efficient photocatalysts for visible‐light‐driven pollutant degradation contributes to sustainable and green solutions to environmental challenges. However, optimizing catalyst composition and structure remains a costly and time‐consuming process. Here, a comprehensive design strategy is presented for the fast development of efficient Al‐doped Mn3O4‐based photocatalysts, combining density functional theory (DFT), machine learning (ML), and laboratory experiments. DFT‐calculated effective mass and bandgaps, serving as indicators of charge mobility and light harvesting, respectively, are employed as descriptors to determine the optimal Al dopant amount. Al0.5Mn2.5O4 is identified as a promising candidate due to its favorable bandgap and charge mobility. To further enhance performance, AlxMn3−xO4/Ag3PO4 heterojunctions are synthesized, leveraging ML to optimize the ratios between AlxMn3−xO4 and Ag3PO4. The best material is determined to be an Al0.5Mn2.5O4/35 wt%‐Ag3PO4 composite, which exhibits a 27‐fold increase in photocatalytic efficiency for methylene blue degradation under visible light compared to pristine Mn3O4. This study not only provided promising photocatalysts for practical pollutant degradation but highlighted the potential of computational and ML‐guided approaches to accelerate photocatalyst discovery. These computational methods provide a framework for the rational design of advanced materials for environmental remediation applications.

Developing efficient and durable electrocatalysts for ethanol electro‐oxidation is crucial for enabling the application of direct ethanol fuel cell technology. Herein, it is demonstrated that Pt–Ga liquid metal‐based nanodroplets can serve as an efficient electrocatalyst to drive ethanol oxidation. The mass activity of Pt is significantly improved by alloying with liquid gallium. Guided by machine learning neural networks, a low‐concentration alkaline electrolyte is specifically formulated to allow electrodes with ultralow Pt loading to demonstrate remarkable activity toward ethanol oxidation with a mass activity as high as 13.47 A mg−1Pt, which is more than 14 times higher than that of commercial Pt/C electrocatalysts (i.e., 0.76 A mg−1Pt). Computational studies reveal that the superior activity is associated with the presence of Ga oxides adjacent to Pt on the catalyst surface which leads to energetically favorable pathways for the oxidation process. The findings reveal untapped opportunities in the realm of liquid metal catalysis and hold great promise for the future development of high‐performance alcohol fuel cells.

The outstanding emission of halide perovskites make them ideal candidates for white emission light-emitting diodes (LEDs) for lighting applications. However, many perovskites contain toxic or scarce elements and have unsatisfactory stability. Here, we report a target-driven approach, based on active learning (AL) techniques, to discover halide perovskites suitable for commercial LED applications. Based on the similarity between halide and oxide perovskites, a model trained on an oxide perovskite dataset plus six AL-selected halide perovskites exhibited excellent performance for photoluminescence quantum yield (PLQY) predictions of oxide and halide perovskites. The model proposed a strong relationship between ionic radii and PLQY, postulated to be due to the self-trap excitons derived from the Jahn-Teller deformation. A novel halide perovskite phosphor, Cs4Zn(Bi0.85Sb0.15)2Cl12:0.01Mn, was designed and synthesized with the aid of the model. It exhibited an 88 % PLQY and outstanding thermal and luminescent stability. A simple white LED was fabricated from this material, exemplifying its commercial potential. This study demonstrates how machine learning techniques can accelerate discovery of next-generation phosphors for high performance single emitter-based white-light emitting devices.

Ionic liquids (ILs) are a diverse class of solvents which can be selected for task-specific properties, making them attractive alternatives to traditional solvents. To tailor ILs for specific biological applications, it is necessary to understand the structure-property relationships of ILs and their interactions with cells. Here, a selection of carboxylate anion-based ILs were investigated as cryoprotectants, which are compounds added to cells before freezing to mitigate lethal freezing damage. The cytotoxicity, cell permeability, thermal behavior, and cryoprotective efficacy of the ILs were assessed with two model mammalian cell lines. We found that the biophysical interactions, including permeability of the ILs, were influenced by considering the IL pair together, rather than as single species acting independently. All of the ILs tested had high cytotoxicity, but ethylammonium acetate demonstrated good cryoprotective efficacy for both cell types tested. These results demonstrate that despite toxicity, ILs may be suitable for certain biological applications. It also demonstrates that more research is required to understand the contribution of ion pairs to structure-property relationships and that knowing the behavior of a single ionic species will not necessarily predict its behavior as part of an IL.

Though platinum (Pt)-based complexes have been recently exploited as immunogenic cell death (ICD) inducers for activating immunotherapy, the effective activation of sufficient immune responses with minimal side effects in deep-seated tumors remains a formidable challenge. Herein, we propose the first example of a near-infrared (NIR) light-activated and lysosomal targeted Pt(II) metallacycle (1) as a supramolecular ICD inducer. 1 synergistically potentiates immunomodulatory response in deep-seated tumors via multiple-regulated approaches, involving NIR light excitation, boosted reactive oxygen species (ROS) generation, good selectivity between normal and tumor cells, and enhanced tumor penetration/retention capabilities. Specifically, 1 has excellent depth-activated ROS production (~7 mm), accompanied by strong anti-diffusion and anti-ROS quenching ability. In vitro experiments demonstrate that 1 exhibits significant cellular uptake and ROS generation in tumor cells as well as respective multicellular tumor spheroids. Based on these advantages, 1 induces a more efficient ICD in an ultralow dose (i.e., 5 μM) compared with the clinical ICD inducer-oxaliplatin (300 μM). In vivo, vaccination experiments further demonstrate that 1 serves as a potent ICD inducer through eliciting CD8+/CD4+ T cell response and Foxp3+ T cell depletion with negligible adverse effects. This study pioneers a promising avenue for safe and effective metal-based ICD agents in immunotherapy.

The rapid growth of resistant microorganisms has caused serious public health issues and poses great pressure on the current healthcare system. In this environment, the necessity of new antibiotic materials is even more prominent. Antimicrobial polymers are a class of polymers that have the ability to eradicate or impede the proliferation of microbes on their surfaces or within their surrounding environment. The mechanism of action of antibacterial polymers also makes them a perfect fit for medical devices. Despite great growing needs, the design of new antibacterial polymers with desired antimicrobial properties is still challenging. In this work, we present the first open-source database for antimicrobial polymers which consists of 489 entries, with 177 unique polymers exhibiting diverse structures and properties. Multiple predictive models were also designed and trained to classify the antimicrobial properties of these polymers. The best-performing random forest model showed an average accuracy of 86.7% in a 10-fold cross-validation test. We also developed multiple guiding pipelines for the design of novel antimicrobial polymers.

Successful additive manufacturing involves the optimisation of numerous process parameters that significantly influence product quality and manufacturing success. One commonly used criteria based on a collection of parameters is the global energy distribution (GED). This parameter encapsulates the energy input onto the surface of a build, and is a function of the laser power, laser scanning speed and laser spot size. This study uses machine learning to develop a model for predicting manufacturing layer height and grain size based on GED constituent process parameters. For both layer height and grain size, an artificial neural network (ANN) reduced error over the data set compared with multi linear regression. Layer height predictions using ANN achieved an R2 of 0.97 and a root mean square error (RMSE) of 0.03 mm, while grain size predictions resulted in an R2 of 0.85 and an RMSE of 9.68 μm. Grain refinement was observed when reducing laser power and increasing laser scanning speed. This observation was successfully replicated in another α + β Ti alloy. The findings and developed models show why reproducibility is difficult when solely considering GED, as each of the constituent parameters influence these individual responses to varying magnitudes.

Though immunogenic cell death (ICD) has garnered significant attention in the realm of anticancer therapies, effectively stimulating strong immune responses with minimal side effects in deep-seated tumors remains challenging. Herein, we introduce a novel self-assembled near-infrared-light-activated ruthenium(II) metallacycle, Ru1105 (λem = 1105 nm), as a first example of a Ru(II) supramolecular ICD inducer. Ru1105 synergistically potentiates immunomodulatory responses and reduces adverse effects in deep-seated tumors through multiple regulated approaches, including NIR-light excitation, increased reactive oxygen species (ROS) generation, selective targeting of tumor cells, precision organelle localization, and improved tumor penetration/retention capabilities. Specifically, Ru1105 demonstrates excellent depth-activated ROS production (∼1 cm), strong resistance to diffusion, and anti-ROS quenching. Moreover, Ru1105 exhibits promising results in cellular uptake and ROS generation in cancer cells and multicellular tumor spheroids. Importantly, Ru1105 induces more efficient ICD in an ultralow dose (10 μM) compared to the conventional anticancer agent, oxaliplatin (300 μM). In vivo experiments further confirm Ru1105's potency as an ICD inducer, eliciting CD8+ T cell responses and depleting Foxp3+ T cells with minimal adverse effects. Our research lays the foundation for the design of secure and exceptionally potent metal-based ICD agents in immunotherapy.

Afterglow imaging plays a crucial role in the cancer treatment field. In contrast to inorganic afterglow imaging agents, organic afterglow imaging agents possess easily modifiable structures and exhibit excellent biocompatibility, thereby presenting significant prospects for application in tumor diagnosis and management. In this review, we summarize the design principles and applications of afterglow probes in tumor imaging and therapy. Finally, we discuss the future challenges and prospects of organic afterglow probes in cancer diagnosis and therapy.

An electrocatalyst with trace vanadium alloyed with liquid metal reduces CO 2 directly into solid carbon.

Aim The urokinase plasminogen activator receptor (uPAR) is a promising biomarker for cancer diagnosis and therapy. We herein fabricated a new type of uPAR-targeted imaging probe Al 18 F-NOTA-VC and preliminarily evaluated its potential application in PET imaging of the glioma model in vivo. Methods Peptide VC was synthesized and identified by MALDI-TOF-MS. The IC 50 between VC/precursor NOTA-VC and uPAR was then determined before the synthesis and purification of Al 18 F-NOTA-VC, followed by further studies of in-vitro properties of Al 18 F-NOTA-VC. Meanwhile, the AE105-based probe followed a similar procedure in-vitro test. Finally, the PET imaging properties, including uPAR-targeting ability and the metabolism of Al 18 F-NOTA-VC, were investigated. Results The VC and NOTA-VC were obtained successfully and demonstrated a good affinity with uPAR. Followed by Al 18 F labeling successfully, excellent properties, including the serum stability, water solubility, and specificity of Al 18 F-NOTA-VC, were obtained in-vitro test compared with AE105 based probe. An excellent tumor uptake and renal excretion data of Al 18 F-NOTA-VC were acquired from in-vivo U87MG tumor model PET imaging, consistent with the subsequent biodistribution study. Conclusion In addition to the excellent specificity and high tumor/normal tissue contrast for uPAR-targeted PET imaging of U87MG tumor, Al 18 F-NOTA-VC possessed promising clearance ability by renal system route. These excellent properties facilitated Al 18 F-NOTA-VC to be a promising imaging agent for uPAR high-expressing tumors and, thus, provided a paradigm for developing peptide-based probes for uPAR-associated disease diagnosis.

Recently, newly developed carbon-based nanomaterials known as carbon dots (CDs) have generated significant interest in nanomedicine. However, current knowledge regarding CD research in the biomedical field is still lacking. An overview of the most recent development of CDs in biomedical research is given in this review article. Several crucial CD applications, such as biosensing, bioimaging, cancer therapy, and antibacterial applications, are highlighted. Finally, CD-based biomedicine's challenges and future potential are also highlighted to enrich biomedical researchers' knowledge about the potential of CDs and the need for overcoming various technical obstacles.

The therapeutic efficacy of immunotherapy for most solid tumors is unsatisfactory due to the "immune-cold" nature. As a form of lytic pro-inflammatory programmed cell death, pyroptosis can release abundant immunogenic damage-associated molecular patterns to extracellular milieu, exhibiting self-cascade amplifying capacity for cancer immunotherapy. However, the activation of the caspase-3-mediated pyroptosis is difficult because mRNA hypermethylation induces the down-regulated GSDME expression. Herein, an integrated strategy to elicit dual pyroptosis pathways is introduced based on calcium sulfide-based nanoreservoirs (denoted as CSSG). CSSG are degraded in the aggravated acidic tumor microenvironment (TME) triggered by the surface GOx participating oxidation and result in the avalanching generation of H2S and Ca2+, which elevate oxidative pressure and induce mitochondrial respiration inhibition. Moreover, the sudden surge in H2S evokes GSDMD mediated pyroptosis through "DUSP6/ERK/NLRP3/caspase-1" signaling pathway, which synergistically reinforces the performance of the Ca2+ overloading initiating GSDME-mediated pyroptosis. CSSG elicit robust pyroptotic cell death to effectively stimulate tumor immunogenicity, and promote an impressive antitumor immunity with effective elimination of primary and distant tumors. Collectively, this work establishes TME-associated degradable nanomaterials to introduce dual pyroptosis pathways simultaneously and has a promising prospect in optimizing cancer immunotherapy.