Guangyu Wang
Assistant Professor
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (11)
Year:
2026
Year:
2026
Year:
2025
Year:
2025
PURPOSE Cancer remains a leading cause of death worldwide. The growing volume of high-throughput single-cell and spatial transcriptomic data sets—particularly those related to cancer—offers immense opportunities as well as analytical challenges for effective data analysis and interpretation. Large language models (LLMs), pretrained on vast data sets and capable of various biomedical tasks, offer a promising solution. This review explores the application of LLMs in cancer research from both cellular and pathologic perspectives, aiming to showcase their potential in advancing precision oncology. MATERIALS AND METHODS We systematically review current LLMs in analyzing single-cell RNA sequencing, spatial transcriptomic, and histology image data, emphasizing their relevance to cancer biology and translational research. RESULTS A total of 24 LLMs, published or in preprint between 2022 and 2025, were selected for review. In single-cell transcriptomics, LLMs have primarily been used for cell type annotation, batch integration, and drug-response prediction. In spatial transcriptomics, LLMs support multislide and multimodal spatial data integration, gene expression imputation, niche and region label prediction, spatial domain identification, cell-cell communication inference, and marker gene detection. In computational pathology, LLMs have been applied to cancer subtyping, detection of rare malignancies, genomic mutation prediction, image segmentation, as well as cross-modal retrieval. Despite these advances, many models remain underoptimized for cancer-specific applications, highlighting the need for domain-specific fine-tuning and scalable adaptation strategies. CONCLUSION LLMs have the potential to significantly advance cancer research by providing scalable and effective tools for analyzing and interpreting single-cell, spatial transcriptomic, and pathology data. Future efforts should prioritize tailoring these models to cancer-specific contexts to enhance their utility in uncovering disease mechanisms, identifying biomarkers, and informing therapeutic strategies.
Year:
2025
Background Atherothrombosis, the main event leading to acute coronary syndrome (ACS), is strongly linked to disturbed blood flow (d-flow) regions. Although the involvement of the Hippo pathway and its kinases Large Tumor Suppressor Kinase 1and 2 (LATS 1 and 2) in mechanical stress responses is known, the mechanisms by which d-flow simultaneously induces senescence, proliferation, and atherothrombosis remain unclear. Methods The role of endothelial cells (EC)-specific LATS1/2 was examined using EC specific knock-out (EKO) mice in a partial left carotid ligation (PLCL) model. Plaque spatial multi-omics analysis was performed by integrating imaging mass cytometry, sequential immunofluorescence (COMET™), and spatial metabolomics at the single-cell level in human and mouse atherosclerotic plaques. Results In tamoxifen-inducible Lats1 homo( -/- ) / Lats2 homo( -/- ) EC-specific knockout (EKO) mice, deletion of LATS1/2 induced by tamoxifen led to fatal outcomes, characterized by severe systemic edema and markedly increased vascular permeability. In contrast, Lats1 het(+ /- ) / Lats2 homo( -/- ) -EKO mice survived and developed atherothrombotic plaques exhibiting neovascularization even without further additional dietary or genetic intervention. Spatial proteomics analysis revealed that LATS1/2 depletion in ECs triggered a senescence-associated stemness (SAS) phenotype, primarily driven by CD38 upregulation. Complementary spatial metabolomics profiling demonstrated a significant increase in sulfite and taurine within LATS1/2-deficient plaques, indicating lowered sulfite oxidase (SUOX) activity. Mechanistically, CD38 upregulation was found to suppress SUOX expression, induce the reverse mode of mitochondrial complex V, and increase succinate dehydrogenase (SDH) activity along with ATP consumption. Paradoxically, despite ATP depletion, this metabolic disturbance enhanced glutamate metabolism and the tricarboxylic acid (TCA) cycle, sustaining EC proliferation under energetically stressed conditions. The combined effect of LATS1/2 deletion and CD38 activation established a unique EC phenotype defined by increased SAS, leading to proliferation, senescence, and eventual cell death. These pathological processes culminated in the formation of atherothrombotic plaques, which were attenuated by inhibition of CD38. Notably, a similar phenotype—marked by metabolically active ECs—was observed in human atherothrombotic plaques, suggesting translational relevance. Conclusion Loss of LATS1/2 in ECs induces SAS state that promotes excessive EC proliferation, senescent cell accumulation, and the development of structurally fragile, leaky neo vessels—hallmarks of atherothrombotic lesions. CD38-mediated SUOX deficiency further amplifies this pathological process by inducing mitochondrial dysfunction, depleting ATP, and triggering compensatory upregulation of glutamate and TCA cycle metabolism. These findings identify a novel LATS1/2–CD38–SUOX axis in ECs that orchestrates SAS-driven atherothrombosis. Targeting CD38 may represent a promising therapeutic strategy to mitigate vascular dysfunction and plaque instability in high-risk ACS patients. Graphical abstract Under normal physiological conditions, LATS1/2 and Lamin A work together to suppress CD38 expression. Lamin A binds directly to the CD38 promoter to repress transcription, and LATS1/2 interact with Lamin A to reinforce this suppression. This collaboration helps maintain low CD38 activity and preserves cellular NAD⁺ levels. However, under disturbed flow (d-flow), both LATS1/2 and Lamin A are downregulated. The loss of this dual repression leads to increased CD38 NADase expression. Elevated CD38 accelerates NAD⁺ consumption, causing NAD⁺ depletion—a hallmark of cellular senescence. Reduced NAD⁺ disrupts key metabolic and stress-response pathways, contributing to the onset of the senescent state. At the same time, CD38 suppresses sulfite oxidase (SUOX), leading to sulfite accumulation and mitochondri
Year:
2025
Artificial intelligence has revolutionized computational biology. Recent developments in omics technologies, including single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics (ST), provide detailed genomic data alongside tissue histology. However, current computational models focus on either omics or image analysis, lacking their integration. To address this, we developed OmiCLIP, a visual-omics foundation model linking hematoxylin and eosin (H&E) images and transcriptomics using tissue patches from Visium data. We transformed transcriptomic data into “sentences” by concatenating top-expressed gene symbols from each patch. We curated a dataset of 2.2 million paired tissue images and transcriptomic data across 32 organs to train OmiCLIP integrating histology and transcriptomics. Building on OmiCLIP, our Loki platform offers five key functions: tissue alignment, annotation via bulk RNA-seq or marker genes, cell type decomposition, image–transcriptomics retrieval, and ST gene expression prediction from H&E images. Compared with 22 state-of-the-art models on 5 simulations, 19 public, and 4 in-house experimental datasets, Loki demonstrated consistent accuracy and robustness.
Year:
2025
Spatial transcriptomics integrates transcriptomics data with histological tissue images, offering deeper insights into cellular organization and molecular functions. However, existing computational platforms mainly focus on genomic analysis, leaving a gap in the seamless integration of genomic and image analysis. To address this, we introduce Thor, a comprehensive computational platform for multi-modal analysis of spatial transcriptomics and histological images. Thor utilizes an anti-shrinking Markov diffusion method to infer single-cell spatial transcriptomes from spot-level data, effectively integrating cell morphology with spatial transcriptomics. The platform features 10 modules designed for cell-level genomic and image analysis. Additionally, we present Mjolnir, a web-based tool for interactive tissue analysis using vivid gigapixel images that display information on histology, gene expression, pathway enrichment, and immune response. Thor's accuracy was validated through simulations and ISH, MERFISH, Xenium, and Stereo-seq datasets. To demonstrate its versatility, we applied Thor for joint genomic-histology analysis across various datasets. In in-house heart failure patient samples, Thor identified a regenerative signature in heart failure, with protein presence confirmed in blood vessels through immunofluorescence staining. Thor also revealed the layered structure of the mouse olfactory bulb, performed unbiased screening of breast cancer hallmarks, elucidated the heterogeneity of immune responses, and annotated fibrotic regions in multiple heart failure zones using a semi-supervised approach. Furthermore, Thor imputed high-resolution spatial transcriptomics data in an in-house bladder cancer sample sequenced using Visium HD, uncovering stronger spatial patterns that align more closely with histology. Bridging the gap between genomic and image analysis in spatial biology, Thor offers a powerful tool for comprehensive cellular and molecular analysis.
Year:
2025
Year:
2024
Exercise changes the tumor microenvironment by remodeling blood vessels and increasing infiltration by cytotoxic immune cells. The mechanisms driving these changes remain unclear. Herein, we demonstrate that exercise normalizes tumor vasculature and upregulates endothelial expression of VCAM1 in YUMMER 1.7 and B16F10 murine models of melanoma but differentially regulates tumor growth, hypoxia, and the immune response. We found that exercise suppressed tumor growth and increased CD8+ T-cell infiltration in YUMMER but not in B16F10 tumors. Single-cell RNA sequencing and flow cytometry revealed exercise modulated the number and phenotype of tumor-infiltrating CD8+ T cells and myeloid cells. Specifically, exercise caused a phenotypic shift in the tumor-associated macrophage population and increased the expression of major histocompatibility complex class II transcripts. We further demonstrated that ERK5 S496A knock-in mice, which are phosphorylation deficient at the S496 residue, ‘mimicked’ the exercise effect when unexercised, yet when exercised, these mice displayed a reversal in the effect of exercise on tumor growth and macrophage polarization compared to wild-type mice. Taken together, our results reveal tumor-specific differences in the immune response to exercise and show that ERK5 signaling via the S496 residue plays a crucial role in exercise-induced tumor microenvironment changes.
Year:
2024
Year:
2023
Collaborators (29)
Paul S. Brookes
Professor
University of Rochester
Keith Youker
Assistant Professor
Baylor College of Medicine
Emily LaVoy
Associate Professor
University of Houston
Keri Schadler
Associate Professor
The University of Texas MD Anderson Cancer Center
Li Lai
Assistant Professor, Research Scientist
Houston Methodist Research Institute
Pengzhi Zhang
Houston Methodist
Riccardo Ballarò
Senior Research Scientist
The University of Texas MD Anderson Cancer Center
Iqbal Mahmud
Assistant Professor
Lamar University
Efstratios Koutroumpakis
Assistant Professor
The University of Texas MD Anderson Cancer Center
Jennifer L. McQuade
The University of Texas MD Anderson Cancer Center
Kristopher Brannan
Assistant Professor
Houston Methodist Research Institute
Nordin M.J. Hanssen
Amsterdam University Medical Centers
Trisha Roy
Vascular Surgeon-Scientist
Houston Methodist Hospital
John Cooke
Professor and Department Chair
Houston Methodist Research Institute
Jing Su
Associate Professor
Indiana University School of Medicine
Ying H. Shen
Baylor College of Medicine
Jun-ichi Abe
The University of Texas MD Anderson Cancer Center
Qin Ma
Vice Chair for Research
Ohio State University
Nicolas Palaskas
Assistant Professor
The University of Texas MD Anderson Cancer Center
sivareddy kotla
Assistant Professor
The University of Texas MD Anderson Cancer Center
Qianqian Song
Associate Professor
Purdue University
Shengyu Li
Assistant Professor
Houston Methodist
Ken Chen
Professor
The University of Texas MD Anderson Cancer Center
Masaki Imanishi
Associate Professor
Tokushima University
James F. Martin
Professor
Baylor College of Medicine
Anita Deswal
Department Chair
The University of Texas MD Anderson Cancer Center
Jared Burks
Professor
The University of Texas MD Anderson Cancer Center
Longhou Fang
Associate Professor
Houston Methodist Research Institute
Erin Seeley
Mass Spectrometry Imaging Facility Director
University of Texas at Austin
Make your next move.
Prepare your application
Review drafts before you send them.