ZWZeyu Wang

Zeyu Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Zeyu Wang has a total of 48 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Directed Scaffold Evolution of Chrysin for the Development of Potent Carboxylesterase 2 Inhibitors

    by Xiao-Ze Bao, Yang Yang, Ci-Qin Li, Jing Zhang, Ze-Yu Wang, Jiangwei Pan, Tianyi Jiang, Guishun Bai, Yue Wei, Zhikun Yang, et al.

    The direct modification of natural products to generate novel scaffolds can greatly enrich the diversity of natural products related chemical space. Herein, we revealed a directed scaffold evolution of natural products (DSENP) strategy, featuring the combination of molecular property prediction and synthetic methodology. Following this, an enantioselective [3+2] heterocyclization of chrysin with chloronitroalkenes was developed to evolve the scaffold of chrysin. The pharmacology studies of the corresponding derivatives towards carboxylesterase 2 identified compounds 3g and 3s as potent and selective CES2 inhibitors. Further in vitro and in vivo studies showed that 3g could inhibit the adipogenesis in cells greatly and ameliorates HFD-induced obesity and lipid metabolic disorder in mice. Moreover, the configuration of stereogenic centers and the nitro group were essential for the potent inhibitory activity. These results indicated that the DSENP should be a promising strategy for the discovery of lead compounds.

  • Fine-grained mutation operators for community hiding using genetic algorithms

    by Shanqing Yu, Jintao Zhou, Meng Zhou, Yidan Song, Jiaxiang Li, Zeyu Wang, Qi Xuan, Silu Mu, Xiaolei Qian

  • Expert-Guided Substructure Information Bottleneck for Molecular Property Prediction

    by Tianyi Jiang, Qiang Yao, Zeyu Wang, Xiaoze Bao, Shanqing Yu, Qi Xuan

    Molecular property prediction plays a crucial role in cheminformatics, yet existing methods are constrained by data scarcity and molecular structural heterogeneity. The Mixture of Experts (MoE) framework adopts a divide-and-conquer approach by partitioning the input space and employing expert models. However, current methods primarily rely on scaffold or atomic-level information, often neglecting fine-grained features such as functional groups. Moreover, existing MoE models lack effective mechanisms to filter redundant and noisy information, limiting prediction accuracy and generalization. To address these challenges, we propose a novel Expert-Guided Substructure Information Bottleneck (ESIB-Mol) framework that integrates MoE learning with the Information Bottleneck (IB) principle to optimize molecular representation learning. ESIB-Mol employs substructure-specific experts to focus on key molecular scaffolds and functional groups, which play a crucial role in determining molecular properties such as bioactivity and pharmacokinetics. Meanwhile, the IB principle is leveraged to filter out redundant and irrelevant information, thereby enhancing prediction accuracy and interpretability. Additionally, a dynamic gating mechanism adaptively assigns molecules to the most relevant expert, optimizing computational efficiency. Extensive experiments on benchmark data sets demonstrate the effectiveness of ESIB-Mol, highlighting its superior performance in molecular property prediction.


  • Organocatalytic Diastereo- and Enantioselective Synthesis of Chlorinated and Trifluoromethylated Pyrrolidines Featuring a Removable Nitro Group

    by Xiaoze Bao, Mengyuan Wu, Guishun Bai, Xusheng Duan, Jiangwei Pan, Yang Yang, Zeyu Wang, Xuanrong Sun, Jean Rodriguez, Damien Bonne, et al.

    A new concept was reported to use nitro as a removable activating group for the stereoselective construction of chlorinated stereogenic centers. This was proven by the enantioselective (3 + 2) cycloaddition of isatin-derived 1,3-dipoles and chloronitroalkenes, affording a series of spiro-oxindoles featuring four consecutive stereogenic centers including one that is chlorinated.


  • Exploring the vulnerability of clustering coefficient: in maximization perspective

    by Xiudi Yan, Bo Zhou, Zeyu Wang, Shanqing Yu, Qi Xuan

    The clustering coefficient (CC), a key measure of network aggregation, plays a crucial role in various fields. However, in studies of network vulnerability, most attention has been directed toward the robustness and connectivity of the overall network, with relatively little focus on the vulnerability of the CC-a limitation that overlooks critical local factors. Moreover, existing research on CC is often hindered by the high computational complexity involved in its calculation. To address these challenges, this paper introduces a novel method for analyzing CC variations and an algorithm specifically designed to maximize the CC, offering new insights into the study of CC vulnerability. The proposed approach reduces computational complexity by leveraging the structural properties of triangles and triplets, which are integral to CC calculation. At the same time, the attack strategy adopts a triangle-closure priority approach, prioritizing edge connections that increase the density of triangular structures. Extensive experiments were conducted on 36 unweighted and weighted networks, comparing our methods with four baseline approaches. The results show that the proposed method achieves excellent clustering coefficient variation with a small attack budget and low time complexity, particularly, the improvisation of CC exceeds 90× in some datasets. These prove that the methods have strong target specificity, and are more feasible and concealed in practical applications.


  • Open Access

    ReStNet: A Reusable & Stitchable Network for Dynamic Adaptation on IoT Devices

    by Maoyu Wang, Yao Lu, Jiaqi Nie, Zeyu Wang, Yun Lin, Qi Xuan, Guan Gui

    With the rapid development of deep learning, a growing number of pre-trained models have been publicly available. However, deploying these fixed models in real-world IoT applications is challenging because different devices possess heterogeneous computational and memory resources, making it impossible to deploy a single model across all platforms. Although traditional compression methods, such as pruning, quantization, and knowledge distillation, can improve efficiency, they become inflexible once applied and cannot adapt to changing resource constraints. To address these issues, we propose ReStNet, a Reusable and Stitchable Network that dynamically constructs a hybrid network by stitching two pre-trained models together. Implementing ReStNet requires addressing several key challenges, including how to select the optimal stitching points, determine the stitching order of the two pre-trained models, and choose an effective fine-tuning strategy. To systematically address these challenges and adapt to varying resource constraints, ReStNet determines the stitching point by calculating layer-wise similarity via Centered Kernel Alignment (CKA). It then constructs the hybrid model by retaining early layers from a larger-capacity model and appending deeper layers from a smaller one. To facilitate efficient deployment, only the stitching layer is fine-tuned. This design enables rapid adaptation to changing budgets while fully leveraging available resources. Moreover, ReStNet supports both homogeneous (CNN-CNN, Transformer-Transformer) and heterogeneous (CNN-Transformer) stitching, allowing to combine different model families flexibly. Extensive experiments on multiple benchmarks demonstrate that ReStNet achieve flexible accuracy-efficiency trade-offs at runtime while significantly reducing training cost.

  • Knowledge-enhanced Relation Graph and Task Sampling for few-shot molecular property prediction

    by Zeyu Wang, Tianyi Jiang, Yao Lu, Xiaoze Bao, Shanqing Yu, Bin Wei, Qi Xuan, Hong Wang


  • Molecular Connectivity Index-Based Data Augmentation for Molecular Property Prediction

    by Zeyu Wang, Tianyi Jiang, Jinhuan Wang, Jiafei Shao, Bin Wei, Qi Xuan, Hong Wang

    Recent years have seen a rapid growth of machine learning in cheminformatics problems. In order to tackle the problem of insufficient training data in reality, more and more researchers pay attention to data augmentation technology. However, few researchers pay attention to the problem of construction rules and domain information of data, which will directly impact the quality of augmented data and the augmentation performance. While in graph-based molecular research, the molecular connectivity index, as a critical topological index, can directly or indirectly reflect the topology-based physicochemical properties and biological activities. In this paper, we propose a novel data augmentation technique that modifies the topology of the molecular graph to generate augmented data with the same molecular connectivity index as the original data. The molecular connectivity index combined with data augmentation technology helps to retain more topology-based molecular properties information and generate more reliable data. Furthermore, we adopt five benchmark datasets to test our proposed models, and the results indicate that the augmented data generated based on important molecular topology features can effectively improve the prediction accuracy of molecular properties, which also provides a new perspective on data augmentation in cheminformatics studies.


  • Label-Flipping Attacks in GNN-Based Federated Learning

    by Shanqing Yu, Jie Shen, Shaocong Xu, Jinhuan Wang, Zeyu Wang, Qi Xuan

    Federated learning offers multi-party collaborative training but also poses several potential security risks. These security issues have been studied more extensively in the context of basic image models, but it is relatively less explored in the field of graphs. Compared to various existing graph-based attack methods, the label-flipping attack does not need to change the graph structure and it is highly stealthy. Therefore, this paper explores a Graph Federated Label Flipping Attack (Graph-FLFA) and proposes a new malicious gradient computation strategy for federated graph models. The goal of this attack method is to maximally disrupt the classification results of specific nodes in the node classification task, without affecting the classification performance of other nodes. This strategy exhibits strong specificity and stealthiness, effectively balancing the influence of various labels and ensuring significant attack effects even when the poisoning ratio is very low. Extensive experiments on four benchmark datasets demonstrate that Graph-FLFA has a high attack success rate in different GNN-based models, achieving the most advanced attack performance. Furthermore, it has the capability to evade detection methods employed in defensive measures.