Zhen Liu’s research while affiliated with East China University of Science and Technology and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (178)


Deep Learning Enhanced Deciphering of Brain Activity Maps for Discovery of Therapeutics for Brain Disorders
  • Article

June 2025

·

3 Reads

iScience

Xianrui Zhang

·

Zhen Liu

·

Xuan Luo

·

[...]

·

Peng Shi


(a) Topography of the study area and (b) the drainage network distribution.
Design rainfall hyetograph (a) Mode I; (b) Mode II; (c) Mode III; (d) Mode IV; (e) Mode V; (f) Mode VI; (g) Mode VII.
Schematic of the model coupling.
Optimization‐simulation framework.
Comparison of simulated and measured runoff discharge processes at (a) Outfall O2 and (b) Outfall O4 during the rainfall on 14 July 2017 after model calibration, and at Outfall O3 during the rainfall on (c) 1 September 2017, and (d) 4 October 2017.

+7

An Optimization‐Simulation Method for Low‐Impact‐Development (LID) Facilities Based on CCMO Algorithm Combining an Integrated Finite Volume Coastal Ocean and Drainage Pipe Model
  • Article
  • Full-text available

May 2025

·

23 Reads

The employment of Low Impact Development (LID) facilities is an effective means to alleviate urban flood in the context of climate change and urbanization. Existing methods for evaluating the hydrological reduction and control effect of LID facilities are mostly based on hydrological models, which have inherent shortages in accurate flood process simulation. In this study, a fully‐distributed bidirectional‐coupled hydrodynamic model based on the Finite Volume Coastal Ocean Model (FVCOM) and a one‐dimensional drainage pipe model is used to evaluate the effectiveness of LID facilities combining the Coevolutionary Constrained Multi‐objective Optimization (CCMO) algorithm. The proposed method is applied to investigate the responses of LID facilities effectiveness in hydrological reduction and control to rainfall patterns in the Yuelai New City, Chongqing, China. Results show that the overflow volume and peak overflow under unimodal rainfall are larger than those under other rainfall patterns. The time lag between initial and peak overflow under unimodal rainfall is the shortest, and it is shorter under small rainfall return period. The flood reduction and delay effect of LID facilities under unimodal rainfall is lower than that under bimodal rainfall, and it is the best under uniform rainfall. At low costs, the peak flood reduction effect of combined LID facilities may be lower than that of a single LID facility with strong permeability, but as costs continue to increase, combined LID facilities show its superiority. The proposed optimization‐simulation method is of great significance for environmental managers to seek best solutions in urban flood control.

Download



Fig. 1 Identification of double-positive cells with GBM and MSCs markers. A t-SNE map displaying the clustering of 3459 cells. Each dot represents an individual cell, colored according to its cluster. B The CopyKat algorithm was used to distinguish between benign (diploid) and malignant (aneuploid) cells, presented in a t-SNE plot. C Bar plots illustrating the mRNA expression levels of the indicated markers, measured by log2-TPM (transcripts per million reads). Each bar represents a single cell, colored by the cell types identified in (A). D A t-SNE plot showing the distribution of double-positive (double-pos) cells, colored by cell types. E IF images showing the cell co-expressing GFAP (red), CD90 (orange), CD73 (yellow), and CD105 (green) in GBM specimens (GBM506). Scale bar: 20 μm (left), 5 μm (right).
Fig. 4 CSF1 exhibits high expression and decreased m 6 A modification levels in hybrids. A Volcano plot showing the distribution of differences in m 6 A modification levels between U87MG and Hybrid86. B Heatmap displaying the top 10 genes upregulation or downregulation in expression among those with decreased m 6 A levels in Hybrid86 compared to U87MG. C Western blotting analysis of CSF1 protein expression in GBM cells and hybrids. D Secreted CSF1 in the supernatants of GBM cells and hybrids were detected by ELISA. Examination of the effect of FTO knockdown on CSF1 expression in hybrids using qPCR (E), western blotting (F), and ELISA (G).
Fig. 8 Schematic diagram illustrating the fusion of GA-MSCs with glioma cells promoting macrophage recruitment and M2 polarization via FTO-mediated CSF1 secretion in the TME.
Fusion of glioma-associated mesenchymal stem/stromal cells with glioma cells promotes macrophage recruitment and M2 polarization via m6A modification of CSF1

Cell Death and Disease

Malignant glioma is the most common primary malignant tumor of the brain in adults, with glioblastoma (GBM) being the most aggressive subtype. Mesenchymal stem/stromal cells (MSCs) have been shown to fuse with tumor cells in various cancers including glioma, thereby regulating tumor progression. However, there has been no systematic research on the fusion of glioma-associated MSCs (GA-MSCs) with glioma cells. Here, it is shown that GA-MSCs are able to spontaneously fuse with glioma cells both in vitro and in vivo. The hybrid cells display significantly lower levels of N6-methyladenosine (m ⁶ A) modification and can modulate the glioma microenvironment by attracting and inducing M2-like polarization of macrophages. Mechanistically, the demethylase fat mass and obesity-associated protein (FTO) mediates demethylation in hybrids and promotes macrophage colony-stimulating factor (CSF1) secretion by increasing its RNA stability in an m ⁶ A-YTH domain family 2 (YTHDF2)-dependent manner. Our study reveals a novel crosstalk mechanism between glioma cells, GA-MSCs, and macrophages in glioma microenvironment, offering potential new approaches for glioma therapy.


Integrating deep learning and molecular dynamics simulations for FXR antagonist discovery

April 2025

·

16 Reads

·

1 Citation

Molecular Diversity

Farnesoid X receptor (FXR) is a key regulator of bile acid, lipid, and glucose homeostasis, making it a promising target for treating metabolic diseases. FXR antagonists have shown therapeutic potential in cholestasis, metabolic disorders, and certain cancers, while clinically approved FXR antagonists remain unavailable and underrepresented in current treatment strategies. To address this, we developed deep learning models for predicting FXR antagonistic activity (ANTCL) and toxicity (TOXCL). Screening 217,345 compounds from the HMDB database identified eleven human metabolite candidates with significant FXR binding potential. Molecular dynamics simulations and binding free energy calculations revealed five more stable complexes compared to the reference compound Gly-MCA, with HMDB0253354 (Fulvestrant) and HMDB0242367 (ZM 189154) standing out for their binding free energies. Hydrophobic interactions, particularly involving residues MET328, PHE329, and ALA291, contributed to their stability. These results demonstrate the effectiveness of deep learning in FXR antagonist discovery and highlight the potential of HMDB0253354 and HMDB0242367 as promising candidates for metabolic disease treatment. Graphical abstract Developed deep learning models for predicting FXR antagonism and toxicity, screening 11 human metabolites with significant FXR binding potential.



Figure 6. PDCL3 promoted vasculogenic mimicry in GA-MSCs thereby accelerating tumor progression in vivo (A) In vivo evaluation of tumorigenesis in orthotopic xenograft nude mice bearing shPDCL3-MSC +GSC and NC-MSC+GSC. Representative bioluminescence images and H&E staining images results were shown (scale bar, 1,000 μm). (B) Immunohistochemistry staining of α-SMA and CD31 in xenograft model nude mice (scale bar, 500 and 100 μm). (C) Immunofluorescence staining of α-SMA and CD31 in xenograft model nude mice (scale bar, 100 μm).
PDCL3 promotes the vasculogenic mimicry of glioma-associated mesenchymal stem cells through VEGFR-2 in glioma environment

April 2025

·

2 Reads

iScience

Glioma-associated mesenchymal stem cells (GA-MSCs), as a critical component of the glioma microenvironment, play an essential role in the development and invasion of glioma. Recent studies have shown that GA-MSCs can differentiate into pericytes under certain conditions, thereby promoting angiogenesis. However, the biological properties and molecular mechanisms underlying this process remain poorly understood. In the glioma microenvironment, our data revealed that the upregulation of Phosducin-like 3 (PDCL3) in GA-MSCs influences the expression of the downstream Vascular Endothelial Growth Factor Receptor 2 (VEGFR-2) and pericyte markers, indicating a transformational relationship between GA-MSCs and pericytes. Furthermore, the vascular mimicry ability of co-cultured GA-MSCs and Human Umbilical Vein Endothelial Cells (HUVECs) is also significantly affected. Glioma induces the upregulation of PDCL3 expression in GA-MSCs, facilitating their transformation of GA-MSCs into pericytes and promoting tumor angiogenesis. These findings suggest that PDCL3 in GA-MSCs could serve as a potential target for anti-angiogenesis therapy in glioma.



Citations (52)


... Anthraquinones exhibit π-π stacking and a variety of molecular states in extraction solutions due to their π-π conjugated system, planar structure, and strong van der Waals interaction between molecules [24][25][26]. It is unidentified how ultrasonic power, solute concentration, and pH affect π-π stacking in ethanol solution, despite the fact that nanofiltration concentration can improve π-π stacking. ...

Reference:

Ultrasonic bidirectional regulation of chrysophanol and aurantio-obtusin self-assembly enhanced nanofiltration separated ethanol extract of cassia seed
A novel flocculant based on "happy molecules" for the efficient removal of NSAIDs and NOM complexes: Role of parallel π-π stacking force
  • Citing Article
  • February 2025

Journal of Hazardous Materials

... Ofloxacin (OFL), a third-generation fluoroquinolone antibiotic, is extensively used to treat Gram-negative bacterial infections in the respiratory, genitourinary, and dermal systems due to its broad-spectrum antimicrobial activity and high bioavailability [4]. However, the fluorine atoms in its molecular structure enhance chemical stability, resulting in poor biodegradability, high environmental persistence, significant bioaccumulation potential, and ecological toxicity [5]. The mass production and improper disposal of OFL have led to widespread environmental contamination, with detected concentrations ranging from 0.5 ng/L to 30 mg/L in surface water, groundwater, and medical wastewater worldwide [6,7]. ...

Asymmetrically coordinated Co-N2S1 sites anchored on biochar regulating the free radical activation of peroxymonosulfate for efficient ofloxacin degradation
  • Citing Article
  • February 2025

Chemical Engineering Journal

... Penggunaan simulasi AI didalam arsitektur melibatkan berbagai teknologi, termasuk machine learning seperti algoritma genetik dan reinforcement learning, dimanfaatkan untuk meningkatkan pencahayaan alami, sistem ventilasi, serta insulasi termal pada bangunan, sehingga berdampak pada penurunan penggunaan energi (Saket Maheshwari & Manya Agrawal, 2024). Beberapa model AI yang banyak digunakan dalam simulasi ini meliputi Generative Adversarial Networks (GAN) diterapkan dalam desain arsitektur untuk menyempurnakan alur kerja dengan menciptakan data desain baru melalui proses pelatihan yang bersifat kontradiktif, sehingga mendorong peningkatan dalam proses kreatif (Li et al., 2024). Algoritma ini seperti proses seleksi alam, memungkinkan eksplorasi yang efektif dalam ruang desain yang rumit, suatu hal yang sangat krusial dalam perancangan Teknik (Zou et al., 2024 (Gammal, 2024) Desain generatif yang menggunakan AI dalam BIM tidak hanya membuat proses desain lebih cepat, tetapi juga mendukung pembangunan ramah lingkungan. ...

The Relationship Between Artificial Intelligence (AI) and Building Information Modeling (BIM) Technologies for Sustainable Building in the Context of Smart Cities

... These foreign scholars have deeply explored the "sustainable" potential of HBIM at the technical level, such as remote access to databases, the establishment of cross-cycle management platforms, the display of interactive icons, and the continuous detection and analysis of heritage structures. Zhang Yunan discussed and pointed out this issue in the article "Domestic HBIM Management Platform and Its Followup Business Expansion" [92], and Zhiwei Zhou also addressed it in the article "Driving Sustainable Cultural Heritage Tourism in China through Heritage Building Information Modeling" [93]. ...

Driving Sustainable Cultural Heritage Tourism in China through Heritage Building Information Modeling

... Reprinted from Shengwei et al., Controlling magnetism in 2D CrI3 by electrostatic doping, Nature Nanotechnology 13 (7), 549-553 (2018), with permission from Springer Nature. [185] e) The sample magnetization in bilayer CrI 3 varies with the electric field under external magnetic fields of +0.44T [186] reversible electrical switching between AF and FM states at significantly lower external magnetic fields. [186] Further, (Figure 8i-m,) a colossal and electrically tunable exchange bias (EB) effect in a CrI 3 homotrilayer on a ferroelectric Y-doped HfO 2 (HYO) substrate, achieving a record-high EB field modulation of 0.48 T (ranging from −0.15 T to +0.33 T). ...

Robust Electric‐Field Control of Colossal Exchange Bias in CrI3 Homotrilayer

... The compound that was consistently obtained in the distillation process with various treatments both control and PEF was Benzenethanol (CAS) or phenyl ethyl alcohol. It is a marker in rose essential oil that produces a distinctive aroma of rose flowers in the essential oil produced [38]. One of the largest hydrocarbon compounds in rose essential oil that plays a role in determining the physical characteristics of essential oils is the stearoptenes group, including 1-nonadecene, nanodecene, and octadecene. ...

Comprehensive investigation on non-volatile and volatile flavor compounds in different varieties of rose tea by UPLC-Q-TOF-MS/MS-based metabolomics and GC-IMS, GC-MS
  • Citing Article
  • September 2024

Journal of Food Composition and Analysis

... Despite great efforts to achieve lead optimization, the failure rate of drug development remains very high (~90%) [5,6]. However, overemphasis on drug exposure in plasma and overlooking drug distribution in disease-targeted tissue/normal tissues may mislead the selection of drug candidates for clinical trials [6][7][8][9][10]. ...

Few-Shot Meta-Learning Applied to Whole Brain Activity Maps Improves Systems Neuropharmacology and Drug Discovery

iScience

... Post-traumatic stress disorder (PTSD) and postpartum depression (PPD) represent critical but often overlooked factors in maternal mortality and morbidity (Liu et al., 2021;Shaik et al., 2023). Although the United Nations Sustainable Development Goals (SDGs) highlight the importance of maternal health, there remains a deficiency in specific interventions that focus on the healing environment within maternal wards, especially in lowresource contexts (Feng et al., 2024;Zuber, et al., 2018;Soman et al., 2024). The existing literature on the impact of healing environments in alleviating maternal psychological distress, especially in developing nations, is quite sparse. ...

Sustainable Healing and Therapeutic Design Driven Well-Being in Hospital Environment

... In this work we discuss how blockchain may be used to improve the construction waste management process to meet sustainability objectives. In addition, Liu et al. (2024) argue that smart city technology integration with the innovation of business models is a new technology that would facilitate the city infrastructure and Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 8s (2025) https://internationalpubls.com resource management processes by proposing a new business model using blockchain. ...

Integration of Smart City Technology and Business Model Innovation