Liver cancer manifests as a profoundly heterogeneous malignancy, posing significant challenges in terms of both therapeutic intervention and prognostic evaluation. Given that the liver is the largest metabolic organ, a prognostic risk model grounded in single-cell transcriptome analysis and a metabolic perspective can facilitate precise prevention and treatment strategies for liver cancer. Hence, we identified 11 cell types in a scRNA-seq profile comprising 105,829 cells and found that the metabolic activity of malignant cells increased significantly. Subsequently, a prognostic risk model incorporating tumor heterogeneity, cell interactions, tumor cell metabolism, and differentially expressed genes was established based on eight genes; this model can accurately distinguish the survival outcomes of liver cancer patients and predict the response to immunotherapy. Analyzing the immune status and drug sensitivity of the high- and low-risk groups identified by the model revealed that the high-risk group had more active immune cell status and greater expression of immune checkpoints, indicating potential risks associated with liver cancer-targeted drugs. In summary, this study provides direct evidence for the stratification and precise treatment of liver cancer patients, and is an important step in establishing reliable predictors of treatment efficacy in liver cancer patients.

Neoadjuvant therapy (NAT) has demonstrated considerable effectiveness in treating locally advanced non-small cell lung cancer (NSCLC). Major pathological response (MPR), defined by the proportion of viable tumor cells in the tumor bed assessed through whole slide images (WSIs), is regarded as a crucial marker for predicting patient prognosis. While artificial intelligence (AI)-assisted WSI analysis has proven to be more effective than manual inspection, the intricate nature of analyzing post-NAT NSCLC WSIs still poses substantial clinical and computational challenges. In this study, we introduce a self-supervised learning model named the Multi-Magnifier Transformer Network (MMT-Net), designed for estimating MPR in NSCLC patients with post-NAT. The MMT-Net framework comprises three main components: a CNN backbone pretrained on unannotated pathological images for feature extraction, a multi-attention mechanism for fusing multi-scale features and a transformer layer for final classification. Validation results on the curated NSCLC post-NAT dataset have demonstrated that MMT-Net is more effective than commonly used methods for estimating MPR. Additionally, attention map visualizations show strong agreement with pathologist annotations, and the model outputs significantly correlate with immunohistochemistry markers such as CK and Ki67, confirming the reliability of MMT-Net. The proposed MMT-Net holds promise for improving the efficiency of MPR estimation, further advancing the clinical application of NAT in NSCLC.

Nitrogen (N2), one of the most stable molecules, can be fixed naturally by lightning. However, industrial nitrogen fixation via the Haber–Bosch and Ostwald processes is energy-intensive and exacerbates carbon dioxide (CO2) emissions. Meanwhile, CO2 has reached alarming atmospheric concentrations, rendering its efficient utilization a critical scientific challenge. Herein, we demonstrate that microdroplets generated by a sonic-spray array reactor can generate surface charges without an external electric field, enabling simultaneous activation of N2 and CO2 at the gas-water interfaces to form ammonium (NH4+, 84.3 µM), nitrate (NO3-, 5.7 µM) and formate (HCO2-, 16.5 µM) within 45 minutes. The activation processes of N2 and CO2 exhibits a synergistic enhancement effect at microdroplet interfaces. Mechanistic investigation reveals novel pathways involving the CO2-facilitated N2 activation and the oxygen radical (•O)-mediated N2 activation. Overall, this research offers a sustainable and energy-efficient strategy for green nitrogen fixation and utilizing CO2 from the atmosphere.