Coal-derived hard carbon (HC) is a promising anode material for sodium-ion batteries (SIBs), yet its capacity is severely limited by dense microcrystalline stacking and insufficient closed pores. Here, we propose an activationcrosslinking balance strategy to construct closed-pore-rich coal-based HCs for high-capacity sodium-ion storage. Using a mild Lewis-acidic activator, the carbon matrix is etched to generate open pores, while coordinationdriven oxygen-bridged crosslinking together with chlorine-induced carbon-carbon crosslinking between aromatic units facilitates pore closure, thereby enabling controllable open-to-closed pore evolution and forming a large number of closed pores. As a result of this synergistic regulation, the specific surface area contributed by closed pores increases substantially from 280 to 1890 m(2) g(-1). The resulting microstructure features expanded interlayer spacing for efficient Na+ transport and abundant closed pores for Na+ storage. Consequently, a high specific capacity of 372 mAh g(-)1 with an ultrahigh plateau contribution of 82% is achieved, ranking among the highest reported for coal-derived HC anodes. Moreover, it exhibits excellent cycling stability, with 93% capacity retention after 1500 cycles at 1 A g(-1). This activation-crosslinking balance strategy offers a simple and scalable route to producing high-performance coal-derived HC anodes for SIBs.

Enhancing ion transport kinetics in thick and dense electrodes is essential for developing supercapacitors that simultaneously deliver high energy and power densities. Although interconnected microporous structures have been demonstrated to effectively enhance ion transport kinetics at the particle scale (< 10 μm), it remains a fundamental question whether such structure can promote ion transport kinetics in practical thick electrodes (> 100 μm). Herein, we prepare an interconnected microporous carbon that uniquely integrates ultrahigh specific surface area, high compaction density, exceptional electrical conductivity, and robust crushing strength. Importantly, we show that the intraparticle interconnected micropores provide effective ion-transport pathways throughout the entire thick and dense electrodes. Although inferior to conventional hierarchically porous carbons at low mass loadings, this interconnected microporous carbon demonstrates more favorable capacitive and kinetic performance under high mass loadings, highlighting its potential for practical applications. Under commercial mass loading, the electrode delivers outstanding gravimetric, volumetric, and areal capacitances (182 F g−1, 98 F cm−3, and 1.82 F cm−2 at 1 A g−1) in organic electrolyte, and retains 41% capacitance at 10 A g−1. This work establishes a new design paradigm for advanced carbon electrodes targeting practical high-energy and high-power energy storage.

AIM: To identify metastasis-associated prognostic genes and construct a robust molecular signature for survival prediction in uveal melanoma (UVM) patients. METHODS: Transcriptomic data and clinical information from 80 UVM patients in the Cancer Genome Atlas (TCGA)-UVM cohort and an external Gene Expression Omnibus (GEO) microarray dataset (GSE73652; 8 non-metastatic vs 5 metastatic cases) were analyzed to identify differentially expressed genes (DEGs). Functional enrichment, protein-protein interaction (PPI) network construction, and survival analyses identified seven metastasis- and prognosis-related genes. Their expression was further examined using public single-cell RNA-seq data (GSE139829; 11 tumors). Experimental validation was performed in UVM cell lines (92.1, OMM1, MEL270) and adult retinal pigment epithelial (ARPE-19) cells using quantitative real-time polymerase chain reaction (qRT-PCR) and Western blotting to confirm transcriptomic trends. A LASSO Cox model was applied to construct a metastasis-related risk Score signature. Tumor immune microenvironment characteristics were evaluated via single-sample gene set enrichment analysis (ssGSEA) and ESTIMATE. Somatic mutation and copy number variation (CNV) profiles were also examined. RESULTS: Seven key genes (UBE2T, KIF20A, DLGAP5, KLC3, TPX2, UBE2C, AURKA) were significantly associated with overall survival and used to construct a metastasis-related riskScore signature, which effectively stratified patients into high- and low-risk groups and served as an independent prognostic factor. qRT-PCR and Western blot results confirmed that the expression levels of selected key genes in UVM cell lines showed significant differences compared to ARPE-19 cells, which were largely consistent with the transcriptomic findings. The high-risk group exhibited reduced immune infiltration and stromal activity. Single-cell analysis revealed these genes were predominantly expressed in a tumor cell cluster characterized by BAP1 loss and high metastatic potential. Mutation and CNV analyses further supported the relevance of these genes to UVM progression. CONCLUSION: This study establishes and validates a seven-gene signature associated with metastasis and prognosis in UVM. The findings provide a framework for understanding molecular determinants of tumor progression and immune microenvironment alterations, and may offer guidance for future mechanistic studies and therapeutic exploration.

Tumor heterogeneity and drug resistance limit single-agent therapies, making combination treatments essential. However, traditional screening methods are costly and inefficient. Here, we present DSimSynergy, a graph deep learning framework for predicting drug synergy. It constructs drug similarity networks from biological process and clinical applications, then learns drug representations through graph convolution on these networks. Subsequently, it combines them with graph attention representations of drug molecular fingerprints and cell line gene expressions to predict synergy scores for drug combinations. Comprehensive benchmarking on multiple independent datasets demonstrates that DSimSynergy consistently outperforms state-of-the-art methods. Model interpretability analysis revealed key genes and pathways underlying drug synergy, while validation on clinical patient and cohort data demonstrated good clinical translational potential and discovered the molecular mechanisms by which drugs generate synergistic effects through “pathway complementary networks”. DSimSynergy efficiently identifies synergistic combinations, reducing experimental costs while elucidating biological mechanisms to overcome resistance and guide personalized treatment.

Immunotherapy has revolutionized cancer treatment, yet substantial inter-patient response heterogeneity limits therapeutic benefit to specific patient subsets. Here, we present PathTIGR, a pathway topology-informed graph representation learning framework that systematically integrates biological pathway network topology knowledge with genome variation information for immunotherapy response prediction. PathTIGR uses a three-component design: (i) pathway graph encoder with multihead attention embeding pathway topology knowledge and cancer genomic variants to pathway representation, (ii) transformer module capturing pathway regulatory dependencies, and (iii) multilayer perceptron synthesizing pathway-level representations to predict immunotherapy response. This architecture enables PathTIGR to capture complex molecular interactions underlying immunotherapy response. Comprehensive validation across multiple independent immunotherapy cohorts demonstrates that PathTIGR achieves superior predictive performance compared to established biomarkers and state-of-the-art deep learning approaches while maintaining biological interpretability through identification of key signatures underlying response heterogeneity. PathTIGR represents an interpretable graph-based learning framework that enhances immunotherapy response prediction and elucidates molecular determinants of therapeutic efficacy, thereby facilitating the advancement of precision cancer immunotherapy.

Silicon (Si)-based anode with ultrahigh theoretical capacity and low lithiation potential has shown great potential for application in solid-state batteries (SSBs). However, the adverse dynamic interface instability and chemical incompatibility lead to the rapid capacity fade of Si-based SSBs. Here, we construct a mechanochemical dual-functional interface by the in-situ elastic plastic-crystal polymer electrolyte (PPE) to enhance the structural integrity and electrochemical stability of Si-based SSBs. The dual-functional interface integrates the energy-dissipative dynamic adaptive interface and LiF-dominated solid-electrolyte interphase (SEI), effectively alleviating adverse mechanical stress accumulation, preserving the interface stability, and facilitating the homogeneous transmission of Li-ions. The interfacial design enables the Si anode to achieve excellent cycling stability with a capacity of similar to 1032 mAh g(-1) after 400 cycles. The NCM811 vertical bar PPE vertical bar Si full cells also demonstrate excellent electrochemical performance with 87.8% capacity retention over 200 cycles at 0.5C. Furthermore, the full cells also demonstrate excellent safety and a wide operating temperature range, enabling stable cycling from 0 to 60 degrees C. This interfacial strategy offers a practical pathway toward Si-based SSBs.

Silicon (Si) anodes are highly promising for high-energy-density solid-state batteries (SSBs), but substantial volume changes during cycling cause persistent solid electrolyte interphase (SEI) fracture and an unstable electrode interface. This challenge is exacerbated in solid-state batteries, where rigid, immobile interfaces present poor mechanical buffering. Herein, an ingenious "mortise-tenon" structural SEI is built by introducing the cyclotetrasiloxane into polymer electrolytes to promote the precise spatial reconfiguration of SEI, achieving an interlock between the cyclotetrasiloxane and LiF-rich inorganic phase, which ensures robust adhesion and structural stability of the SEI under large volume changes. The resulting Si||Li half cells deliver a high capacity of 1553.6 mAh g-1 at a high current density of 12 A g-1. The NCM811||Si full cells achieve a high-capacity retention of 97.6% at 0.5 C, showing only 0.12% capacity decay per cycle over 300 cycles. The LFP||Si full cells also show a low decay rate of 0.07‰ per cycle across 700 cycles. This strategy provides stable SEI engineering for the practical application of high-energy-density Si-based SSBs.

Hard carbon (HC) is a promising anode for sodium‐ion batteries (SIBs), but coal‐derived HCs often exhibit low reversible capacity and poor initial Coulombic efficiency (ICE) due to irreversible sodium (Na) loss on defective carbons. Presodiation can directly improve ICE, but the imprecise and slow process can lead to under‐ or oversodiation and the formation of thick and unstable byproducts. Here, we propose a high precision and fast presodiation by aryl‐sodium (Ar–Na) compounds dissolved in tetrahydrofuran (THF) with controlled potential and Ar–Na binding energy. Based on the thermodynamic driving force (redox potential) and ionic transfer kinetics (Ar–Na binding strength), a dual‐descriptor design principle for presodiation agent is established. Phenanthrene–sodium (Ph–Na) with a moderate ionic binding energy of − 0.92 eV and a matched redox potential of 0.24 V versus Na⁺/Na enables ∼100% ICE within 60 s and facilitates the formation of an ultrathin inorganic‐rich SEI in the battery that enhances interfacial kinetics and cycling stability. The presodiated HC delivers a reversible capacity of 308.9 mAh g −1 , and paired with a Na 3 V 2 (PO 4 ) 3 (NVP) cathode exhibits 94.8% ICE, 99.2 mAh g −1 discharge capacity, and 82.6% capacity retention after 350 cycles, demonstrating a scalable presodiation strategy for practical SIBs.

Rechargeable metal-chlorine (Li/Na-Cl2) batteries potentially have a high energy density, but the significant amount of electrolyte consumed to produce active metal chlorides for reversible chlorine conversion severely limits their real electrochemical performance. Herein, we use a cathode with precast metal chloride in the graphene layers as the initial active material to save the sacrificial electrolyte and deliver a fundamentally different start-up operation mode for metal-chlorine batteries. Furthermore, the metal chloride confined by the graphene layers achieves in situ confining conversion with gaseous chlorine during long cycling, causing substantially improved cathode kinetics in a lean electrolyte. With this cathode, both Li/Na-Cl2 batteries demonstrate higher capacity and prolonged cycling performance. Typically, the obtained Na-Cl2 batteries could deliver a high areal capacity (3 mAh cm-2) and stable life over 300 cycles under lean electrolyte conditions (20-60 μL). This work demonstrates the practical significance of utilizing a graphene interlayer to confine metal chloride as an initial active material for rechargeable alkali-metal-Cl2 batteries.

Although a high stack pressure (≥50 MPa) enhances solid-solid contacts in solid-state batteries (SSBs), it poses impracticality for commercialization. This work proposes a self-pressure silicon (Si)-carbon composite anode that enables stable operation under reduced external pressure (≤2 MPa). The self-pressure anode features a prestress structure that can effectively alleviate the internal and external stress simultaneously, which is fabricated with ionic-conductive poly(ethylene oxide) (PEO)/lithium salt-coated carbon nanotubes (CNTs) being compressed by shrinking graphene hydrogel. The capillary-driven hydrogel shrinkage generates internal pressure, compensating for the volumetric expansion (up to 300%) of Si. This creates dynamic solid-solid interfaces between compressed CNTs/PEO and expanding Si, ensuring both mechanical stability and ion/electron transport. The SSBs with this self-pressure anode have a long cycle life of 700 cycles and a high capacity retention of 79.2% in an organic/inorganic composite electrolyte without external pressure (0 MPa). The half-cell using a sulfide solid-state electrolyte reached 700 cycles and was able to achieve a stable cycle life at the lowest 2 MPa stack pressure. This design resolves interfacial challenges by prestress in SSBs.

The mainstream graphite anode in lithium batteries encounters obstacles including the capacity reduction and polarization during fast-charging, which is mainly restricted by the sluggish ion transfer kinetics. Here, a universal electronegative interfacial modification strategy is proposed for improving the fast-charging performance of the graphite (Gr) anode, in which the electronegative-COOH groups on the graphite/electrolyte interface serve as a Li+ reservoir by the pre-adsorption of a large amount of Li+ through electrostatic interaction. The electron-rich interface and the enrichment of Li+ at the interface weaken the Li+-solvent interaction and provide a larger Li+ potential difference, thereby accelerating desolvation process and inducing the formation of inorganic-rich SEI. Therefore, the interfacial kinetics of the graphite anode was enhanced substantially by the weakened Li+-solvent interaction and LiF-rich SEI interface. As a result, the Gr@rGO anode with optimized-COOH groups and appropriate defects achieved a great rate performance of 192 mAh g-1 at 4C, small polarization of 0.026 V, and excellent cycling stability with 97 % after 600 cycles in pouch cells. In addition, universal application of-COOH modified interphase for fast-charging performance in high-energy-density anode systems, including SiO and nano-silicon anodes, was proved to be feasible, revealing the effectiveness of the electronegative regulation for fast-charging lithium-ion battery anodes.

Silicon (Si)-based solid-state battery operation under low stack pressure (<= 2 MPa) is very challenging owing to the poor interfacial compatibility and stability. In this work, we show an interfacial depassivation design for a solid-state battery assembled by nano-Si anodes and polyvinylidene fluoride (PVDF)-based composite electrolytes to eliminate the need for stack pressure. We first demonstrate the interfacial passivation of Si particles plays a critical role in impacting the battery performance, which is caused by the decomposition of the residual solvent and lithium salt transported from the PVDF electrolyte. Here, a solid electrolyte interphase immobilization strategy enabled by polyaniline on Si adsorbing the solvated lithium salt is implemented to achieve interfacial depassivation and enhance the interfacial transport kinetics. As a result, a significantly prolonged 500-cycle life and ultralow capacity decay of 0.0094 %/cycle for a Si-based full solid-state cell operating free of external pressure is obtained. This work provides an effective interface engineering for practical solid-state batteries without additional stack pressure.

A micropore confinement and fusion strategy is proposed to eliminate the interfacial mismatch and achieve molecular-level integrated catalysis interfaces for long-life all-solid-state Li–S batteries.

Accurate characterization of cellular states is the foundation for precise prediction of drug sensitivity in cancer cell lines, which in turn is fundamental to realizing precision oncology. However, current deep learning approaches have limitations in characterizing cellular states. They rely solely on isolated genetic markers, overlooking the complex regulatory networks and cellular mechanisms that underlie drug responses. To address this limitation, this work proposes DeepCCDS, a Deep learning framework for Cancer Cell Drug Sensitivity prediction through Characterizing Cancer Driver Signals. DeepCCDS incorporates a prior knowledge network to characterize cancer driver signals, building upon the self-supervised neural network framework. The signals can reflect key mechanisms influencing cancer cell development and drug response, enhancing the model's predictive performance and interpretability. DeepCCDS has demonstrated superior performance in predicting drug sensitivity compared to previous state-of-the-art approaches across multiple datasets. Benefiting from integrating prior knowledge, DeepCCDS exhibits powerful feature representation capabilities and interpretability. Based on these feature representations, we have identified embedding features that could potentially be used for drug screening in new indications. Further, this work demonstrates the applicability of DeepCCDS on solid tumor samples from The Cancer Genome Atlas. This work believes integrating DeepCCDS into clinical decision-making processes can potentially improve the selection of personalized treatment strategies for cancer patients.

Immunotherapy has revolutionized cancer treatment, but predicting patient response remains challenging. Herein, we present iPRISM (Intelligent Predicting Response to cancer Immunotherapy through Systematic Modeling), which is a novel network‐based model that integrates multiomics data to predict immunotherapy outcomes. In this approach, iPRISM incorporates gene expression, biological functional network, tumor microenvironment characteristics, immune‐related pathways, and clinical data to provide a comprehensive view of factors influencing immunotherapy efficacy. Using stepwise logistic regression, we identified key predictive features and validated iPRISM across multiple cohorts including melanoma, bladder cancer, non‐small cell lung cancer, and stomach adenocarcinoma. We also find that iPRISM outperforms the existing methods, achieving high predictive accuracy and demonstrating significant prognostic value for overall and progression‐free survival. By identifying key genetic and immunological factors, this model provides a new insight for more personalized treatment strategies and combination therapies to overcome resistance mechanisms. iPRISM can be accessed at CRAN: https://CRAN.R‐project.org/package=iPRISM.

The development of hard carbon (HC) anodes with the low-cost coal precursor for sodium-ion batteries (SIBs) is usually limited by sluggish kinetics and low capacity. The main reason is that the abundant aromatic frameworks in coal produce over-stacked microcrystalline during high-temperature carbonization, restricting sodium ion intercalation and diffusion. Here, we propose a molecular templating and doping strategy using phenylboronic acid (PhB) to regulate the microcrystalline structure of coal-derived HCs. The π-π interaction between PhB and coal aromatic suppresses the excessive stacking, while boron (B) doping perturbs charge distribution and introduces intra-domain defects. These effects lead to the formation of twisted turbostratic domains with an expanded interlayer spacing, enabling high ionic and electronic conductivity for sodium ion storage. The reduced surface electronegativity by B-doping also favors the formation of a stable anion-derived interface. As a result, the optimized HC delivers a high reversible capacity of 321 mAh g −1 at 50 mA g −1 and maintains a capacity of 249 mAh g −1 at the high current density of 2.5 A g −1 . This work demonstrates that molecular templating offers an effective route to balance capacity and reaction kinetics in coal-based HC anodes.

Rechargeable lithium-chlorine (Li-Cl2) batteries are recognized as powerful candidates for energy storage due to high energy density and adaptability in harsh environments. However, porous materials used in Li-Cl2 battery cathodes exhibit poor confinement of active chlorine species, which significantly challenges the rechargeability of batteries. Herein, we propose a blocking pore design to enable strong confinement and highly reversible conversion of chlorine species, where a narrow pore-entrance blocking effect confines LiCl growth in the pore body and suppresses Cl2 spillage from the pores. As a result, the obtained Li-Cl2 battery demonstrates a high gravimetric capacity of 3863 mAh g-1, along with a super-long cycle life (750 cycles) and remarkable rate capability (616 mAh g-1 at 30 A g-1). This work presents a novel chlorine confinement mechanism to improve the reversibility of chlorine conversion in metal-chlorine batteries.

Micro-silicon (mSi)-based anodes have garnered considerable interest due to their potential for high energy density and lower-cost energy storage systems. However, the large volume changes during repetitive lithiation and delithiation can lead to severe fracture and pulverization of mSi particles, ultimately resulting in rapid performance degradation. In this work, we address this problem by doping mSi with germanium (Ge), which improves the electrical conductivity, increases lattice spacing, and optimizes Li+ diffusion channels. This method can achieve an initial coulombic efficiency of up to 95% and accelerated reaction kinetics for mSi-based anodes. Specifically, the Si20Ge anode exhibits a reversible capacity of 1109.3 mAh g-1 at the current density of 4 A g-1 after 150 cycles. When integrated into PVDF-based solid-state full cells, the Si20Ge anode, paired with a LiNi0.8Co0.1Mn0.1 (NCM811) cathode, retains 83% of its capacity after 200 cycles at the current density of 1C. This work offers valuable insights into the rational structural design of mSi alloyed anode materials for achieving higher-performance Li-ion batteries.

In silico drug prioritization may be a promising and time-saving strategy to identify potential drugs, standing as a faster and more cost-effective approach than de novo approaches. In recent years, artificial intelligence has greatly evolved the drug development process. Here, we present a novel computational framework for drug prioritization, labyrinth, designed to simulate human knowledge retrieval and inference to identify potential drug candidates for each disease. With the integration of up-to-date clinical trials, literature co-occurrences, drug-target interactions, and disease similarities, our framework achieves over 90% predictive accuracy across clinical trial phases and strong alignment with clinical practice in TCGA cohorts. We have demonstrated effectiveness across 20 different disease categories with robust ROC-AUC metrics and the balance between predictive accuracy and model interpretability. We further demonstrate its effectiveness at both the population and the individual levels. This study not only demonstrates the capacity for its drug prioritization but underscores the importance of aligning computational models with intuitive human reasoning. We have wrapped the core function into an R package named labyrinth, which is freely available on GitHub under the GPL-v2 license (https://github.com/hanjunwei-lab/labyrinth).

Cancer originates from dysregulated cell proliferation driven by driver gene mutations. Despite numerous algorithms developed to identify genomic mutational signatures, they often suffer from high computational complexity and limited clinical applicability. Here, we presented ProgModule, an advanced computational framework designed to identify mutation driver modules for cancer prognosis and immunotherapy response prediction. In ProgModule, we introduced the Prognosis-Related Mutually Exclusive Mutation (PRMEM) score, which optimizes the balance between exclusive mutation coverage and the incorporation of mutation combination mechanisms critical for cancer prognosis. Applying to BLCA and HNSC cohorts, ProgModule successfully identified driver modules that stratify patients into distinct prognostic subgroups, and the combination of these modules could serve as an effective prognostic biomarker. Extending our method to diverse cancers, ProgModule presented robust prognostic performance and stability across model parameters, including stopping criteria and network topology. Moreover, our analysis suggested that driver modules can predict immunotherapeutic benefit more effectively than existing signatures. Further analyses based on published CRISPR data indicated that genes within these modules may serve as potential therapeutic targets. Altogether, ProgModule emerges as a powerful tool for identifying mutation driver modules as prognostic and immunotherapy response biomarkers, and genes within these modules may be used as potential therapeutic targets for cancer, offering new insights into precision oncology.