Skip to content
Log in
Sponsored Have an interesting project or opportunity? → Hire me ↗
Sponsored Artificial Intelligence: A Modern Approach, Global Edition Artificial Intelligence: A Modern Approach, Global Edition

Author

Wencai Ye

View on Semantic Scholar

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Open access Aug 2026

maA dual-target modulator of STK11 and PP2Acα from the flowers of Chrysanthemum indicum mitigates drug-induced liver injury.

INTRODUCTION Hormesis, characterized by low-dose stimulation and high-dose inhibition, is a biphasic regulatory phenomenon, and its underlying mechanisms remain elusive. Drug-induced liver injury (DILI) can progress to liver fibrosis, liver failure, and ultimately death, and natural products hold considerable promise for the treatment of DILI. OBJECTIVES To identify the active constituents and underlying biphasic regulatory mechanism of Chrysanthemum indicum against DILI. METHODS Structural elucidation of the new compounds was achieved through integrated interpretation of HRESIMS, 1D and 2D NMR, and ECD. Signaling pathway was determined by mitochondrial transplantation in vitro and in vivo and RNA sequencing. Target proteins were validated by the drug affinity responsive target stability-mass spectrometry analyses, isothermal titration calorimetry, cellular thermal shift assay, shRNA, and liver-specific knockdown mice. Protein sites were validated by truncation experiments, molecular dynamics simulations, and point mutations. RESULTS A total of 21 guaianolide sesquiterpenoids, including 13 new ones, were isolated and identified from the flowers of C. indicum. Interestingly, the new compound chrysanthemolide I (CI) alleviated acetaminophen-induced liver injury in vitro and in vivo. Mitochondria isolated from CI-treated hepatocytes attenuated DILI. CI attenuated AMP-activated protein kinase (AMPK)-mediated mitochondrial oxidative stress while enhancing AMPK-dependent mitochondrial biogenesis and mitophagy. At low doses, CI binds directly to ALA-205 and ARG-301 of serine/threonine kinase 11 (STK11) with high affinity to activate AMPK; at medium doses, binding of CI to STK11 reaches saturation, leading to peak AMPK activity; at high doses, CI additionally binds to SER-261 of serine/threonine-protein phosphatase 2A catalytic subunit α isoform (PP2Acα) with low affinity to inhibit AMPK activation. Furthermore, liver-specific knockdown of both STK11 and PP2Acα largely diminished the protective effect of CI against DILI. CONCLUSION Novel guaianolide sesquiterpenoid CI was identified as an affinity-dependent dual-target regulator of STK11 and PP2Acα to alleviate DILI.

Fei Zhou, Yu Liu, Haoyu Zhao et al. · 0 citations
Aug 2026

Mol-ME: Enhancing Molecular Property Prediction via Multi-Modal Alignment Learning and Ensemble Methods.

Molecular deep learning plays an important role in addressing challenging molecular property prediction tasks. However, labeled molecular data remain scarce, and the majority of existing studies predominantly employ single-modal methods. Most single-modal models face limitations in simultaneously capturing molecular topological features and modeling long-range dependencies in sequences. In this study, we propose a novel multimodal alignment framework for joint modeling of molecular graphs and sequences, called Mol-ME. The framework incorporates a data augmentation strategy to enhance model performance under limited labeling conditions. Mol-ME comprises four core modules. The first module consists of dual encoders that generate graph-based and sequence-based molecular representations, which are then aligned through contrastive learning. The second module, a gated cross-modal fusion network, enables fine-grained integration of these representations by leveraging both the cross-attention mechanism and the gating mechanism. The third module is a motif-aware feature extractor that captures latent relationships among molecular substructures. The final module employs ensemble learning to predict on extracted representations, which captures complex nonlinear relationships and compensates for the modeling limitations of single shallow networks. Experimental results on 9 benchmark data sets demonstrate that Mol-ME consistently outperforms all baseline methods, achieving new state-of-the-art (SOTA) performance in molecular property prediction.

Baoren Huang, Mu Chen, Junjie Luo et al. · 0 citations