Researchers have created graphene-based membranes that mimic the chemical gating effect in biological membranes, achieving record ion selectivity (>10,000) between monovalent and bivalent metal ions including Li⁺ and Mg²⁺ ions. The membrane allows adaptive ion permeation in response to certain signaling ions like Al³⁺, thus enables smart and ultrafast ion separation for applications in filtration, sensing, and energy technologies.

Monitoring the vitality of shellfish during cold storage is critical for ensuring seafood quality, yet conventional methods remain invasive, subjective, and slow. 

Chinese researchers identified histone demethylase LSD1 as a critical guardian of ovarian reserve. When deleted in mouse oocytes, LSD1 deficiency disrupted mitochondrial homeostasis and triggered ferroptosis, causing massive loss of dormant primordial follicles. This discovery explains how dormant follicles ("second wave") are maintained differently from active follicles ("first wave") and provides a foundation for treating premature ovarian insufficiency.

A low-field anomalous quantum oscillation is discovered at ultra-clean LaAlO3/SrTiO3 interfaces, which is attributed to the time-reversal symmetry-protected transport along quasi-1D ferroelastic domain walls.

A breakthrough study has developed a model that accurately predicts the failure of lithium metal anodes using electrochemical curve data from just the first two cycles. The model offers new insights into failure mechanisms and provides a predictive tool that could speed up the development of more reliable lithium metal batteries, which are critical for high-energy applications.

ChemELLM, a 70-billion-parameter LLM tailored for chemical engineering, outperforms leading LLMs (e.g., Deepseek-R1) on ChemEBench across 101 tasks, trained on ChemEData’s 19 billion pretraining and 1 billion fine-tuning tokens, accelerating lab-to-fab innovation.

The utilization of covalent organic frameworks (COFs) holds great potential for achieving tailorable tuning of catalytic performance through bottom-up modulation of the reticular structure. In this work, we show that a single-point structural alteration in the linkage within a nickel phthalocyanine (NiPc)-based series effectively modulates the catalytic performance of the COFs in electrochemical CO₂ reduction reaction (CO₂RR). A NiPc-based COF series with three members which possess the same NiPc unit but different linkages, including piperazine, dioxin, and dithiine, have been constructed by nucleophilic aromatic substitution reaction between octafluorophthalocyanine nickel and tetrasubstituted benzene linkers with different bridging groups. Among these COFs, the dioxin-linked COF showed the best activity of CO₂RR with a current density of CO (j_CO) =  − 27.99 mA cm⁻² at − 1.0 V (versus reversible hydrogen electrode, RHE), while the COF with piperazine linkage demonstrated an excellent selectivity of Faradaic efficiency for CO (FE_CO) up to 90.7% at a pretty low overpotential of 0.39 V. In addition, both a high FE_CO value close to 100% and a reasonable j_CO of − 8.20 mA cm^–2 at the potential of − 0.8 V (versus RHE) were obtained by the piperazine-linked COF, making it one of the most competitive candidates among COF-based materials. Mechanistic studies exhibited that single-point structural alteration could tailor the electron density in Ni sites and alter the interaction between the active sites and the key intermediates adsorbed and desorbed, thereby tuning the electrochemical performance during CO₂RR process.

As part of the Healthy China Action, the prevention and treatment of hemophilia has attracted widespread attention in China. This article systematically summarizes the current status and progress of hemophilia in the country in terms of prevalence, diagnostic technology, prevention methods, and treatment plans.

Recent advances in spatial omics and single-cell omics have significantly reshaped biomarker discovery in tumor immunotherapy by addressing critical challenges posed such as tumor heterogeneity, immune evasion, and variability within the tumor microenvironment (TME).

While immunotherapeutic strategies—such as immune checkpoint inhibitors and adoptive T-cell transfer—have demonstrated promising clinical outcomes, their effectiveness is hindered by low response rates and immune-related adverse events (irAEs). Thus, identifying reliable biomarkers is essential for predicting treatment efficacy, minimizing irAEs, and facilitating patient stratification. Spatial omics integrates molecular profiling with spatial localization, providing comprehensive insights into the cellular organization and functional states of the TME. By revealing spatial patterns of immune cell infiltration and tumor heterogeneity, this approach enhances our predictive capacity for therapeutic response. Similarly, single-cell omics yields high-resolution analysis of cellular heterogeneity, capturing transcriptomic, epigenomic, and metabolic signatures at the single-cell level. The combined application of spatial and single-cell omics has led to the identification of previously undetected biomarkers, including rare immune cell subsets implicated in resistance mechanisms. Beyond spatial transcriptomics (ST), this technological landscape also includes spatial proteomics and metabolomics, which further facilitate the study of dynamic tumor–immune interactions. Multi-omics integration offers a comprehensive overview of biomarker landscapes, while the rapid evolution artificial intelligence approaches enhances the analysis of complex, multidimensional datasets—ultimately enhancing predictive power and clinical utility. Despite substantial progress, challenges remain in standardization, data integration, and real-time monitoring. Nevertheless, incorporating spatial omics and single-cell omics into biomarker research holds transformative potential for personalized cancer immunotherapy. These emerging strategies pave the way for innovative diagnostic and therapeutic interventions, enabling precision oncology and elevating treatment outcomes for patients a wide range of tumor profiles.

This review aims to provide a comprehensive summary of the integration of spatial omics and single-cell omics in tumor immunotherapy biomarker discovery. Specifically, it focuses on how these emerging technologies address challenges related to tumor heterogeneity, immune evasion, and the dynamic TME. By elaborating on the principles, applications, and clinical potential of these technologies, the review will also critically evaluate their limitations, challenges, and the current gaps in their translational applications.