Yuxi Liu
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
Research Subjects
Yuxi Liu has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Jiang Bian
Indiana University School of Medicine
12 shared publications Zhenhao Zhang
North West Agriculture and Forestry University
10 shared publications- Shaowen Qin
Flinders University
4 shared publications Song Wang
Department of Electrical and Computer Engineering University of Virginia Charlottesville Virginia USA
4 shared publications- Yu Huang
Indiana University
4 shared publications Jiang Bian
3 shared publications
Latest Publications
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 12964-12973
GatorTrio: Topology-Refined Tri-View Graph Learning for Spatial Domain Identification in Spatial Transcriptomics
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 12964-12973
Open AccessGatorTrio: Topology-Refined Tri-View Graph Learning for Spatial Domain Identification in Spatial Transcriptomics
- Open Access
GatorPrism: Prototype-Conditioned Routing across Coalition Graph Experts for Spatial Multi-Omics Integration
Spatial multi-omics integration requires balancing cross-omics consensus with modality-specific signals whose importance varies across tissue locations. We present GatorPrism, a self-supervised coalition graph mixture-of-experts framework that explicitly separates cross-omics consensus from modality-specific structure and adaptively integrates their contributions across tissue locations. A joint expert encodes an intersection-based consensus graph, while modality-private experts capture mixed spatial-molecular structures; a prototype-conditioned router assigns spot-specific coalition weights. GatorPrism is trained end-to-end to preserve shared and modality-specific neighborhoods, align co-registered modalities, maintain spatial coherence, and prevent routing collapse. Across eight spatial multi-omics datasets, GatorPrism achieved strong performance across nine clustering metrics against nine competing methods. In human tonsil, inferred domains were supported by concordant RNA and ADT markers and distinct functional programs. In embryonic mouse brain, routing profiles revealed anatomically localized shared, RNA-private, and ATAC-private states supported by transcriptomic, chromatin-accessibility, and motif evidence. These results establish GatorPrism as an accurate and interpretable framework that reveals how shared and modality-specific molecular signals organize tissue structure. Source code is available at https://github.com/Gator-Group/GatorPrism.
Open AccessGatorPrism: Prototype-Conditioned Routing across Coalition Graph Experts for Spatial Multi-Omics Integration
Spatial multi-omics integration requires balancing cross-omics consensus with modality-specific signals whose importance varies across tissue locations. We present GatorPrism, a self-supervised coalition graph mixture-of-experts framework that explicitly separates cross-omics consensus from modality-specific structure and adaptively integrates their contributions across tissue locations. A joint expert encodes an intersection-based consensus graph, while modality-private experts capture mixed spatial-molecular structures; a prototype-conditioned router assigns spot-specific coalition weights. GatorPrism is trained end-to-end to preserve shared and modality-specific neighborhoods, align co-registered modalities, maintain spatial coherence, and prevent routing collapse. Across eight spatial multi-omics datasets, GatorPrism achieved strong performance across nine clustering metrics against nine competing methods. In human tonsil, inferred domains were supported by concordant RNA and ADT markers and distinct functional programs. In embryonic mouse brain, routing profiles revealed anatomically localized shared, RNA-private, and ATAC-private states supported by transcriptomic, chromatin-accessibility, and motif evidence. These results establish GatorPrism as an accurate and interpretable framework that reveals how shared and modality-specific molecular signals organize tissue structure. Source code is available at https://github.com/Gator-Group/GatorPrism.
- Open Access
GatorDuo: Global-Consistency Dual-Graph Refinement With Pseudo-Label Agreement for Spatial Transcriptomics
Spatial transcriptomics (ST) measures gene expression together with spatial coordinates, enabling spatial domain identification of coherent tissue regions. Many recent approaches rely on graph-based modeling to combine spatial neighborhoods and transcriptomic (gene-expression) similarity, yet neighborhood construction is often unreliable under sparsity and technical noise. As a result, spurious cross-domain shortcut edges can persist in static graphs and propagate misleading signals during message passing, ultimately blurring domain boundaries and weakening cluster separability. In this paper, we propose GatorDuo, a topology-aware dual-graph contrastive self-supervised framework for robust spatial domain identification that couples gene-expression similarity with spatial proximity through complementary neighborhood graphs. GatorDuo introduces global-consistency–based graph refinement that uses a pseudo-label agreement mask to suppress cross-domain shortcut edges in both views, thus stabilizing neighborhood topology for representation learning. To avoid manual tuning of domain resolution, GatorDuo further employs a contextual bandit reinforcement-learning strategy to adaptively select the clustering granularity (the number of clusters) used for refinement. The refined view-specific embeddings are integrated via a hybrid-routing Mixture-of-Experts (MoE) module to generate a unified embedding, optimized with contrastive objectives augmented by an MoE-alignment term. Across eight public benchmarks spanning sequencing- and imaging-based ST at spot and single-cell resolution, and compared with ten representative baselines, GatorDuo consistently delivers strong and robust spatial domain identification performance across multiple clustering metrics, while yielding informative unified embeddings that can support downstream biological analyses.
Open AccessGatorDuo: Global-Consistency Dual-Graph Refinement With Pseudo-Label Agreement for Spatial Transcriptomics
Spatial transcriptomics (ST) measures gene expression together with spatial coordinates, enabling spatial domain identification of coherent tissue regions. Many recent approaches rely on graph-based modeling to combine spatial neighborhoods and transcriptomic (gene-expression) similarity, yet neighborhood construction is often unreliable under sparsity and technical noise. As a result, spurious cross-domain shortcut edges can persist in static graphs and propagate misleading signals during message passing, ultimately blurring domain boundaries and weakening cluster separability. In this paper, we propose GatorDuo, a topology-aware dual-graph contrastive self-supervised framework for robust spatial domain identification that couples gene-expression similarity with spatial proximity through complementary neighborhood graphs. GatorDuo introduces global-consistency–based graph refinement that uses a pseudo-label agreement mask to suppress cross-domain shortcut edges in both views, thus stabilizing neighborhood topology for representation learning. To avoid manual tuning of domain resolution, GatorDuo further employs a contextual bandit reinforcement-learning strategy to adaptively select the clustering granularity (the number of clusters) used for refinement. The refined view-specific embeddings are integrated via a hybrid-routing Mixture-of-Experts (MoE) module to generate a unified embedding, optimized with contrastive objectives augmented by an MoE-alignment term. Across eight public benchmarks spanning sequencing- and imaging-based ST at spot and single-cell resolution, and compared with ten representative baselines, GatorDuo consistently delivers strong and robust spatial domain identification performance across multiple clustering metrics, while yielding informative unified embeddings that can support downstream biological analyses.
- Open Access
- 8 April 2026
- Published by Wiley in Small Methods
- Vol. 10 (9) , e00006
GatorST: A Versatile Contrastive Meta‐Learning Framework for Spatial Transcriptomic Data Analysis
Recent advances in spatial transcriptomics (ST) have revolutionized the understanding of cellular functions by providing gene expression profiles with rich spatial context. Effectively learning spatial representations is essential for downstream analyses and requires robust integration of spatial and transcriptomic information. Although existing methods show promise, they often fail to capture local (neighbor‐level) and global (tissue‐wide) contexts, and methods based on contrastive learning often rely on augmentation strategies that introduce noise and instability. GatorST, a novel and versatile framework, explicitly integrates graph‐based modeling with advanced meta‐learning strategies to generate spatially informed representations of ST data. Locally, a spot‐spot graph connects each node to its nearest neighbors, while two‐hop subgraphs capture fine‐grained spatial context. Globally, gene expression profiles are clustered to produce pseudo‐labels, providing weak supervision for representation learning. An episodic training strategy inspired by meta‐learning further enhances GatorST's ability to generalize to new spatial contexts, ensuring robust integration of local and global spatial information. Comprehensive comparisons with fifteen state‐of‐the‐art methods demonstrate that GatorST consistently outperforms existing approaches in identifying spatial domains, imputing gene expression, removing batch effects, and inferring spatial trajectories. By integrating local spatial topology with global gene expression patterns, GatorST provides biologically meaningful representations that advance key downstream analyses.
- 8 April 2026
- Published by Wiley in Small Methods
- Vol. 10 (9) , e00006
Open AccessGatorST: A Versatile Contrastive Meta‐Learning Framework for Spatial Transcriptomic Data Analysis
Recent advances in spatial transcriptomics (ST) have revolutionized the understanding of cellular functions by providing gene expression profiles with rich spatial context. Effectively learning spatial representations is essential for downstream analyses and requires robust integration of spatial and transcriptomic information. Although existing methods show promise, they often fail to capture local (neighbor‐level) and global (tissue‐wide) contexts, and methods based on contrastive learning often rely on augmentation strategies that introduce noise and instability. GatorST, a novel and versatile framework, explicitly integrates graph‐based modeling with advanced meta‐learning strategies to generate spatially informed representations of ST data. Locally, a spot‐spot graph connects each node to its nearest neighbors, while two‐hop subgraphs capture fine‐grained spatial context. Globally, gene expression profiles are clustered to produce pseudo‐labels, providing weak supervision for representation learning. An episodic training strategy inspired by meta‐learning further enhances GatorST's ability to generalize to new spatial contexts, ensuring robust integration of local and global spatial information. Comprehensive comparisons with fifteen state‐of‐the‐art methods demonstrate that GatorST consistently outperforms existing approaches in identifying spatial domains, imputing gene expression, removing batch effects, and inferring spatial trajectories. By integrating local spatial topology with global gene expression patterns, GatorST provides biologically meaningful representations that advance key downstream analyses.
- Open Access
GatorSC: Multi-Scale Cell and Gene Graphs with Mixture-of-Experts Fusion for Single-Cell Transcriptomics
Single-cell RNA sequencing (scRNA-seq) enables high-resolution characterization of cellular heterogeneity, but its rich, complementary structure across cells and genes remains underexploited, especially in the presence of technical noise and sparsity. Effectively leveraging this multi-scale structure is essentially an information fusion problem that requires integrating heterogeneous graph-based views of cells and genes into robust low-dimensional representations. In this paper, we introduce GatorSC, a unified representation learning framework that models scRNA-seq data through multi-scale cell and gene graphs and fuses them with a Mixture-of-Experts architecture. GatorSC constructs a global cell–cell graph, a global gene–gene graph, and a local gene–gene graph derived from neighborhood-specific subgraphs, and learns graph neural network embeddings that are adaptively fused by a gating network. To learn noise-robust and structure-preserving embeddings without labels, we couple graph reconstruction and graph contrastive learning in a unified self-supervised objective applied to both cell- and gene-level graphs. We evaluate GatorSC on 19 publicly available scRNA-seq datasets covering diverse tissues, species, and sequencing platforms. Experiments showed that GatorSC consistently outperforms state-of-the-art deep generative, graph-based, and contrastive methods for cell clustering, gene expression imputation, and cell-type annotation. The learned embeddings are used for accurate trajectory inference, recovery of canonical marker gene programs, and cell-type-specific pathway signatures in an Alzheimer’s disease single-nucleus dataset. GatorSC provides a flexible foundation for comprehensive single-cell transcriptomic analysis and can be readily extended to multi-omic and spatial modalities.
Open AccessGatorSC: Multi-Scale Cell and Gene Graphs with Mixture-of-Experts Fusion for Single-Cell Transcriptomics
Single-cell RNA sequencing (scRNA-seq) enables high-resolution characterization of cellular heterogeneity, but its rich, complementary structure across cells and genes remains underexploited, especially in the presence of technical noise and sparsity. Effectively leveraging this multi-scale structure is essentially an information fusion problem that requires integrating heterogeneous graph-based views of cells and genes into robust low-dimensional representations. In this paper, we introduce GatorSC, a unified representation learning framework that models scRNA-seq data through multi-scale cell and gene graphs and fuses them with a Mixture-of-Experts architecture. GatorSC constructs a global cell–cell graph, a global gene–gene graph, and a local gene–gene graph derived from neighborhood-specific subgraphs, and learns graph neural network embeddings that are adaptively fused by a gating network. To learn noise-robust and structure-preserving embeddings without labels, we couple graph reconstruction and graph contrastive learning in a unified self-supervised objective applied to both cell- and gene-level graphs. We evaluate GatorSC on 19 publicly available scRNA-seq datasets covering diverse tissues, species, and sequencing platforms. Experiments showed that GatorSC consistently outperforms state-of-the-art deep generative, graph-based, and contrastive methods for cell clustering, gene expression imputation, and cell-type annotation. The learned embeddings are used for accurate trajectory inference, recovery of canonical marker gene programs, and cell-type-specific pathway signatures in an Alzheimer’s disease single-nucleus dataset. GatorSC provides a flexible foundation for comprehensive single-cell transcriptomic analysis and can be readily extended to multi-omic and spatial modalities.
- 21 July 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 169, 104880
Variational temporal deconfounder network for individualized treatment effect estimation with longitudinal observational data
- 21 July 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 169, 104880
Variational temporal deconfounder network for individualized treatment effect estimation with longitudinal observational data
- 16 July 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 327
A natural language processing-based approach for early detection of heart failure onset using electronic health records
- 16 July 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 327
A natural language processing-based approach for early detection of heart failure onset using electronic health records
- Open Access
GatorST: A Versatile Contrastive Meta-Learning Framework for Spatial Transcriptomic Data Analysis
Introduction: Recent advances in spatial transcriptomics (ST) technologies have revolutionized our understanding of cellular functions by providing gene expression profiles with rich spatial context. Effectively learning spatial representations is crucial for downstream analyses and requires robust integration of spatial information with transcriptomic data. While existing methods have shown promise, they often fail to adequately capture both local (neighbor-level) and global (tissue-wide) spatial contexts. Moreover, they tend to rely heavily on augmentation strategies, which can introduce noise and instability. Objectives: This study aims to introduce and demonstrate a novel, versatile framework called GatorST, which explicitly combines graph-based modeling with advanced learning strategies to generate spatially informed representations of ST data. GatorST is designed to improve various downstream tasks, including identification of spatial domains, gene expression imputation, batch effect removal, and trajectory inference. Methods: GatorST constructs a spot-spot graph by connecting each node to its k nearest spatial neighbors and extracts two-hop neighborhood subgraphs to capture local context. At the global level, gene expression profiles are clustered using soft K-means to generate pseudo-labels, which serve as weak supervision signals within a contrastive learning framework. This process encourages the alignment of embeddings with shared pseudo-labels while separating those with different labels. GatorST further adopts an episodic training strategy inspired by meta-learning, wherein each episode consists of a support set for contrastive optimization and a disjoint query set for embedding classification, guided by the pseudo-labeled data. This design enables the model to classify unseen samples based on learned embeddings, thereby enhancing its generalization to new spatial contexts. Results: Comprehensive comparisons with fifteen state-of-the-art methods across fourteen spatial transcriptomics datasets demonstrate that GatorST consistently achieves superior performance in identifying spatial domains, imputing gene expressions, and removing batch effects. The results showcase the versatility and strong generalization capabilities of GatorST across diverse tissue types and experimental settings. Conclusion: GatorST effectively integrates spatial topology and global gene expression through graph-based modeling, pseudo-labeling, and contrastive meta-learning. This framework generates biologically meaningful representations and significantly improves key downstream tasks, including spatial domain identification, gene expression imputation, batch effect removal, and trajectory inference.
Open AccessGatorST: A Versatile Contrastive Meta-Learning Framework for Spatial Transcriptomic Data Analysis
Introduction: Recent advances in spatial transcriptomics (ST) technologies have revolutionized our understanding of cellular functions by providing gene expression profiles with rich spatial context. Effectively learning spatial representations is crucial for downstream analyses and requires robust integration of spatial information with transcriptomic data. While existing methods have shown promise, they often fail to adequately capture both local (neighbor-level) and global (tissue-wide) spatial contexts. Moreover, they tend to rely heavily on augmentation strategies, which can introduce noise and instability. Objectives: This study aims to introduce and demonstrate a novel, versatile framework called GatorST, which explicitly combines graph-based modeling with advanced learning strategies to generate spatially informed representations of ST data. GatorST is designed to improve various downstream tasks, including identification of spatial domains, gene expression imputation, batch effect removal, and trajectory inference. Methods: GatorST constructs a spot-spot graph by connecting each node to its k nearest spatial neighbors and extracts two-hop neighborhood subgraphs to capture local context. At the global level, gene expression profiles are clustered using soft K-means to generate pseudo-labels, which serve as weak supervision signals within a contrastive learning framework. This process encourages the alignment of embeddings with shared pseudo-labels while separating those with different labels. GatorST further adopts an episodic training strategy inspired by meta-learning, wherein each episode consists of a support set for contrastive optimization and a disjoint query set for embedding classification, guided by the pseudo-labeled data. This design enables the model to classify unseen samples based on learned embeddings, thereby enhancing its generalization to new spatial contexts. Results: Comprehensive comparisons with fifteen state-of-the-art methods across fourteen spatial transcriptomics datasets demonstrate that GatorST consistently achieves superior performance in identifying spatial domains, imputing gene expressions, and removing batch effects. The results showcase the versatility and strong generalization capabilities of GatorST across diverse tissue types and experimental settings. Conclusion: GatorST effectively integrates spatial topology and global gene expression through graph-based modeling, pseudo-labeling, and contrastive meta-learning. This framework generates biologically meaningful representations and significantly improves key downstream tasks, including spatial domain identification, gene expression imputation, batch effect removal, and trajectory inference.
- 2 June 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 168, 104851
GatorCLR: Personalized predictions of patient outcomes on electronic health records using self-supervised contrastive graph representation
- 2 June 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 168, 104851
GatorCLR: Personalized predictions of patient outcomes on electronic health records using self-supervised contrastive graph representation
- 1 May 2025
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 9 (9) , e2400291
Optimizing Strategy for Lung Cancer Screening: From Risk Prediction to Clinical Decision Support
PURPOSE: Low-dose computed tomography (LDCT) screening is effective in reducing lung cancer mortality by detecting the disease at earlier, more treatable stages. However, high false-positive rates and the associated risks of subsequent invasive diagnostic procedures present significant challenges. This study proposes an advanced pipeline that integrates machine learning (ML) and causal inference techniques to optimize lung cancer screening decisions. MATERIALS AND METHODS: Using real-world data from the OneFlorida+ Clinical Research Consortium, we developed ML models to predict individual lung cancer risk and estimate the benefits of LDCT screening. Explainable artificial intelligence techniques were applied to identify key risk factors, ensuring transparency and trust in the model's predictions. Causal ML methods were used to estimate individualized treatment effects of LDCT screening, answering the critical what-if question regarding risk reduction from LDCT. RESULTS: We defined a high-risk cohort of 5,947 patients who underwent LDCT, along with matched controls, to evaluate the framework. Our models demonstrated predictive performance with AUCs of 0.777 and 0.793 for 1-year and 3-year risk predictions, respectively. Causal modeling showed a consistent reduction in lung cancer risk across different subgroups due to LDCT. Specifically, the doubly robust model showed an average risk reduction of 9.5% for males and 12% for females. Age-stratified results indicated a 9.5% reduction for individuals age 50-60 years, a 7.5% reduction for those age 60-70 years, and the largest reduction of 15.1% for the 70-80 age group. CONCLUSION: Integrating ML and causal inference into clinical workflows offers a robust tool for enhancing lung cancer screening. This pipeline provides accurate risk assessments and actionable insights tailored to individuals, empowering clinicians and patients to make informed screening decisions. The differential risk reduction across subgroups highlights the importance of personalized screening in improving outcomes for populations at risk of lung cancer.
- 1 May 2025
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 9 (9) , e2400291
Optimizing Strategy for Lung Cancer Screening: From Risk Prediction to Clinical Decision Support
PURPOSE: Low-dose computed tomography (LDCT) screening is effective in reducing lung cancer mortality by detecting the disease at earlier, more treatable stages. However, high false-positive rates and the associated risks of subsequent invasive diagnostic procedures present significant challenges. This study proposes an advanced pipeline that integrates machine learning (ML) and causal inference techniques to optimize lung cancer screening decisions. MATERIALS AND METHODS: Using real-world data from the OneFlorida+ Clinical Research Consortium, we developed ML models to predict individual lung cancer risk and estimate the benefits of LDCT screening. Explainable artificial intelligence techniques were applied to identify key risk factors, ensuring transparency and trust in the model's predictions. Causal ML methods were used to estimate individualized treatment effects of LDCT screening, answering the critical what-if question regarding risk reduction from LDCT. RESULTS: We defined a high-risk cohort of 5,947 patients who underwent LDCT, along with matched controls, to evaluate the framework. Our models demonstrated predictive performance with AUCs of 0.777 and 0.793 for 1-year and 3-year risk predictions, respectively. Causal modeling showed a consistent reduction in lung cancer risk across different subgroups due to LDCT. Specifically, the doubly robust model showed an average risk reduction of 9.5% for males and 12% for females. Age-stratified results indicated a 9.5% reduction for individuals age 50-60 years, a 7.5% reduction for those age 60-70 years, and the largest reduction of 15.1% for the 70-80 age group. CONCLUSION: Integrating ML and causal inference into clinical workflows offers a robust tool for enhancing lung cancer screening. This pipeline provides accurate risk assessments and actionable insights tailored to individuals, empowering clinicians and patients to make informed screening decisions. The differential risk reduction across subgroups highlights the importance of personalized screening in improving outcomes for populations at risk of lung cancer.