XWXiting Wang

Xiting 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

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xiting Wang has a total of 58 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    TCMCard: A High-Confidence Digital Infrastructure for Traditional Chinese Medicine Quantified by Multi-Dimensional Evidence Integration

    by Yuqi Wang, Wenqing Dong, Junkai Yao, Kai Wang, Leyuan Zhang, Yuanrong Wang, Shanshan Guo, Haorui Li, Hengjia Cai, Xiting Wang, et al.

    Network pharmacology has become a widely used approach for deciphering multi-component, multi-target mechanisms of traditional Chinese medicine (TCM). Here we introduce TCMCard, a high-confidence digital infrastructure built on a Multi-Dimensional Evidence Integration (MDEI) framework. The framework integrates experimental activity data from authoritative chemical databases, literature-derived evidence, and structure-based similarity inference. Preprocessing steps include chemical structure normalization, species-specific filtering, and target quality scoring. Applied to conventional interaction datasets, this pipeline leads to the removal of over 60% of low-confidence noise. TCMCard supports network pharmacology exploration through an interactive visualization platform, and module analysis identifies functionally relevant communities that offer insights into the synergistic actions of TCM formulas. Overall, TCMCard may help move the field beyond simple data aggregation toward evidence-informed curation and quality-driven analysis. As an interactive and publicly accessible platform, it reveals an organized backbone within complex interaction networks, offering a more reliable basis for understanding multi-component synergy in TCM.

  • Open Access

    EndoAtlas: A Multi-Center Clinical Benchmark and Report-Generation Framework for Enhanced Endoscopy Interpretation

    by Qian-Qian Chen, Jiafeng Wang, Yutong Sun, Jihong Liu, Mengni Zhang, Huikai Li, Yan Xia, Meng Liu, Xiting Wang, En-Qiang Ling Hu

    Background: Clinical interpretation of endoscopic reports is often limited by heterogeneous documentation and shortages of specialist expertise. We aimed to develop a multicentre clinical benchmark and a Chinese large language model (LLM) report-generation framework to improve endoscopic report understanding and structured information extraction. Methods: We built EndoAtlas from de-identified electronic health records from five tertiary hospitals in China (N=6161), including surgical reports, pathology reports, and operative diagnoses. We evaluated in-context learning (ICL; 3–10 shots) and low-rank adaptation (LoRA) fine-tuning for report-style generation and structured field extraction. Performance was measured with precision, recall, F1, and BLEU-4. Findings: Increasing ICL demonstrations improved performance across models; for example, Doubao improved from F1 25·84 (3-shot) to 31·77 (10-shot) and BLEU-4 from 57·03 to 69·00, and DeepSeek improved from F1 25·78 to 32·56 with BLEU-4 up to 66·91. In report-generation evaluations, models achieved robust performance for surgical and pathology report generation (F1 64·00–70·95; BLEU-4 77·68–81·62). GLM4 achieved the highest performance for surgical report generation (F1 67·94; BLEU-4 81·41), while DeepSeek achieved the highest performance for pathology report generation (F1 70·95; BLEU-4 81·62). Interpretation: EndoAtlas provides a multicentre, extensible benchmark for endoscopic report understanding and structured report generation. Report-style generation with targeted LLM adaptation could facilitate more standardised endoscopic interpretation and scalable clinical decision assistance, particularly in resource-limited settings.

  • Open Access

    Identification of causal plasma metabolite biomarkers for ischemic stroke using Mendelian randomization and mediation analysis

    by Jian Lyu, Fumei Liu, Yan Chai, Xiting Wang, Yi Liu, Yanming Xie

    The Global Burden of Disease Study 2021 reports that stroke remains a leading cause of death, with ischemic stroke (IS) presenting significant challenges in screening, prevention, and treatment. We explored the causal effects of 1,400 plasma metabolites on IS outcomes using a two-sample Mendelian randomization (MR) framework. We assessed causal relationships between IS and 11 common clinical risk factors and further examined these relationships for metabolites. Mediation analysis identified mechanisms for metabolites affecting both IS and its risk factors. Finally, a phenome-wide association study (PheWAS) MR analysis evaluated the side effects and additional indications of IS-associated metabolites across 3,948 phenotypes from the UKBB GWAS. Nineteen metabolites showed a causal relationship with IS. MR analysis confirmed body mass index (BMI), high-density lipoprotein (HDL), systolic blood pressure (SBP), diastolic blood pressure (DBP), and type 2 diabetes (T2D) as risk factors for IS. Among 136 metabolites associated with at least one IS risk factor, 132 were linked to risk factors but not directly to IS. BMI, DBP, and coffee intake mediated the causal relationship between IS and the levels of 1-stearoyl-GPG (18:0), 1-oleoyl-2-linoleoyl-GPE (18:1/18:2), Octadecadienedioate (C18:2-DC), and X-24,951. Phe-MR analysis indicated that these metabolites were protective and affected other indications similarly to IS. Our findings reveal causal pathways and identify four potential biomarkers for IS, providing new insights for its screening, prevention, and treatment.

  • Open Access

    Artificial intelligence accelerates the identification of nature-derived potent LOXL2 inhibitors

    by Xiaowei Jia, Meng Liu, Yushi Tang, Jingyan Meng, Ruolin Fang, Xiting Wang, Cheng Li

    The role of LOXL2 in cancer has been widely demonstrated, but current therapies targeting LOXL2 are not yet fully developed. We believe that selective nature-derived inhibition of LOXL2 may provide a better therapeutic approach for the treatment of cancer. Therefore, we adopted a comprehensive approach combining deep learning and traditional computer-aided drug design methods to screen LOXL2 selective inhibitors. Bioactivity and affinity of the potential LOXL2 inhibitors were determined by molecular docking and virtual screening. At the same time, we experimentally tested the effect of potential LOXL2 inhibitors on cancer cells. Validation showed that it could inhibit proliferation and migration, promote apoptosis of CT26 cells, and reduce the expression level of LOXL2 protein. As a result, we identified a potent LOXL2 inhibitor: the natural product Forsythoside A, and demonstrated that Forsythoside A has an inhibitory effect on tumors.


  • Open Access

    Exploring the shared biomarkers between cardioembolic stroke and atrial fibrillation by WGCNA and machine learning

    by Jingxin Zhang, Bingbing Zhang, Tengteng Li, Yibo Li, Qi Zhu, Xiting Wang, Tao Lu

    Background: Cardioembolic Stroke (CS) and Atrial Fibrillation (AF) are prevalent diseases that significantly impact the quality of life and impose considerable financial burdens on society. Despite increasing evidence of a significant association between the two diseases, their complex interactions remain inadequately understood. We conducted bioinformatics analysis and employed machine learning techniques to investigate potential shared biomarkers between CS and AF.Methods: We retrieved the CS and AF datasets from the Gene Expression Omnibus (GEO) database and applied Weighted Gene Co-Expression Network Analysis (WGCNA) to develop co-expression networks aimed at identifying pivotal modules. Next, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis on the shared genes within the modules related to CS and AF. The STRING database was used to build a protein-protein interaction (PPI) network, facilitating the discovery of hub genes within the network. Finally, several common used machine learning approaches were applied to construct the clinical predictive model of CS and AF. ROC curve analysis to evaluate the diagnostic value of the identified biomarkers for AF and CS.Results: Functional enrichment analysis indicated that pathways intrinsic to the immune response may be significantly involved in CS and AF. PPI network analysis identified a potential association of 4 key genes with both CS and AF, specifically PIK3R1, ITGAM, FOS, and TLR4.Conclusion: In our study, we utilized WGCNA, PPI network analysis, and machine learning to identify four hub genes significantly associated with CS and AF. Functional annotation outcomes revealed that inherent pathways related to the immune response connected to the recognized genes might could pave the way for further research on the etiological mechanisms and therapeutic targets for CS and AF.


  • Open Access

    Acupuncture for fatigue in parkinson’s disease: A protocol for systematic review and meta-analysis

    by Yibo Li, Jingxin Zhang, Xiaohan Liu, Tengteng Li, Bingbing Zhang, Xiting Wang, Tao Lu

    Background: Parkinson’s disease (PD) is a common movement disorder characterized by bradykinesia, rigidity, and resting tremors. Fatigue is a common disabling symptom but is easily ignored in PD. Half of the PD patients were influenced by fatigue. Acupuncture is one of the conservative treatments for fatigue related to other conditions, especially in China. Therefore, we perform a systematic review and meta-analysis to evaluate the evidence for acupuncture’s effectiveness, safety, and cost benefits for the treatment. Methods: This protocol is based on the previously published randomized controlled trial (RCT) studies. A literature search will be performed on the following database: PubMed, the Cochrane Library, Chinese BioMedical Literature Database, China National Knowledge Infrastructure (CNKI), China Science and Technology Journal Database (VIP), and Wanfang Data. According to the Cochrane Risk of Bias Tool and the level of evidence for results, we will assess the quality of the included studies by using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) method. The Review Manager (v5.3) software will be applied to statistical analysis. Results: From the study, we will assess the effectiveness, safety, and cost-benefit of acupuncture on fatigue relief and functional improvement in patients with Parkinson’s Disease. PROSPERO registration number: CRD42020163155 Strengths and limitations of this study: This paper emphasizes the importance of assessing acupuncture in treating Parkinson’s fatigue symptoms and provides methodological guidance for the evaluation of clinical evidence. This study 1) provides a research protocol, 2) facilitates the reasonable evaluation for acupuncture in the treatment of Parkinson’s fatigue symptoms, and 3) raises the potential importance of acupuncture in Parkinson’s fatigue symptoms. Due to the lack of acupuncture reports, there may be a limitation of the small sample size.

  • Open Access

    Corrigendum to “Network Pharmacology-Based Strategy for Predicting Active Ingredients and Potential Targets of Coptis chinensis Franchin Polycystic Ovary Syndrome”

    by Jian Xiong Ma, Miaoyong Ye, Ke Ma, Kang Zhou, Yingying Zhang, Xiting Wang, Hongxuan Tong

    [This corrects the article DOI: 10.1155/2021/6651307.].

  • Open Access

    Network Pharmacology-Based Strategy for Predicting Active Ingredients and Potential Targets of Coptis chinensis Franchin Polycystic Ovary Syndrome

    by Jian Xiong Ma, Miaoyong Ye, Ke Ma, Kang Zhou, Yingying Zhang, Xiting Wang, Hongxuan Tong

    Background. Polycystic ovary syndrome (PCOS) causes low fertility in females. Coptis chinensis (C. chinensis) is used to clear heat and dampness, purify fire, and detoxify in traditional Chinese medicine (TCM). Although C. chinensis has demonstrated efficacy against PCOS in clinical practice, there are no available data regarding the bioactive components of C. chinensis, their targets, and molecular mechanisms underlying their effects. Methods and Results. Network pharmacology was used to analyze the bioactive components of C. chinensis, their targets, and signaling pathways underlying their effects. The TCM systems pharmacology database and analysis platform (TCMSP) was used to screen 14 effective active ingredients and 218 targets of C. chinensis. The GeneCards, OMIM, and PharmGkb databases were used to screen 3517 disease targets for PCOS, and 102 common targets of drugs and diseases were screened using R Cytoscape that was utilized to build a drug-active ingredient-disease target interaction network, and the STRING platform was utilized to construct a common target protein-protein interaction network, including 102 nodes and 221 edges. Key targets of C. chinensis for the treatment of PCOS included JUN, MAPK, IL6, CXCL8, FOS, and IL1B. A total of 123 gene ontology (GO) terms and 129 pathways were acquired by GO and KEGG enrichment analyses. The AGEs/RAGE, TNF, IL-17, MAPK, and HIF-1 signaling pathways were closely related to PCOS and may be the core pathways involved in PCOS. Schrodinger software was used to evaluate the interaction between active components and their targets and explore binding modes. Furthermore, based on the prediction of network pharmacology study, a mouse model of PCOS was established to evaluate the curative role and underlying mechanisms of C. chinensis. The results showed that C. chinensis treatment reversed histopathological damage of the ovary and also ameliorated the mRNA and protein expression levels of the predicted hub targets (MAPK1, CXCL8, IL-6, and IL-1β). These results indicated that WZYZP has a protective effect on spermatogenesis disorder, suggesting that it could be an alternative choice for male infertility therapy. Conclusions. This preliminary study verified the basic pharmacological effects and mechanisms of C. chinensis, a TCM, in the treatment of PCOS. These results indicate that the therapeutic effects of C. chinensis on PCOS may be achieved by regulating the expression of inflammatory factors. This study provides new insights for the systematic exploration of the mechanism of traditional Chinese medicine.


  • Optimizing Pharmacokinetic Property Prediction Based on Integrated Datasets and a Deep Learning Approach

    by Xiting Wang, Meng Liu, Lan Zhang, Yun Wang, Yu Li, Tao Lu

    Oral bioavailability (OBA)-related pharmacokinetic properties, such as aqueous solubility, lipophilicity and intestinal membrane permeability, play a significant role in drug discovery. However, their measurement is usually costly and time consuming. Therefore, prediction models based on diverse approaches have been established in recent decades. Computational prediction of molecular properties has become an important step in drug discovery, aiming to identify potential drug-like candidates and reduce costs. However, limitations related to dataset capacity and algorithm adaptation still place restrictions on the applicability of the related models. In this study, we considered both dataset and algorithm optimization to address the challenge of predicting OBA-related molecular properties. Benchmark datasets of aqueous solubility (LogS), lipophilicity (LogD) and membrane permeability measured using the Caco-2 cell line (LogPapp) were constructed by merging and calibrating experimental data from diverse articles and databases. Then, a novel molecular property prediction model, called a multiembedding-based synthetic network (MESN), was generated by applying a deep learning algorithm based on the synthesis of multiple types of molecular embeddings. MESN achieves performance improvements over other state-of-the-art methods for the prediction of aqueous solubility, lipophilicity and membrane permeability. Results were also obtained using several other algorithms and independent validation datasets as a control study. Moreover, a dimension reduction analysis (based on t-distributed stochastic neighbor embedding, t-SNE) and an atomic feature similarity analysis showed that the molecular embeddings extracted from the MESN model exhibit good clustering and diversity. Overall, considering the fundamental role of the data and the superior prediction performance of the model, we highlight the applicability of MESN on benchmark datasets for further utility in drug-discovery-related molecular property prediction.