Jeffrey Chuang

Jackson Laboratory
Country flagUnited States

Research Interests

ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

Articles (54)

Year:

2026

Pediatric sarcomas present diagnostic challenges due to their rarity and diverse subtypes, often requiring specialized pathology expertise and costly genetic tests. To overcome these barriers, we developed a computational pipeline leveraging deep learning methods to accurately classify pediatric sarcoma subtypes from digitized histology slides. To ensure classifier generalizability and minimize center-specific artifacts, a dataset comprising 867 whole slide images (WSIs) from three medical centers and the Children's Oncology Group (COG) was collected and harmonized. Multiple convolutional neural network (CNN) and vision transformer (ViT) architectures were systematically evaluated as feature extractors for SAMPLER-based WSI representations, and input parameters, such as tile size combinations and resolutions, were tested and optimized. The analysis showed that advanced ViT foundation models (UNI, CONCH) significantly outperformed earlier approaches, and incorporating multiscale features enhanced classification accuracy. The optimized models achieved high performance, distinguishing rhabdomyosarcoma (RMS) from non-rhabdomyosarcoma (NRSTS) with an AUC of 0.969 and differentiating RMS subtypes (alveolar vs. embryonal) with an AUC of 0.961. Additionally, a two-stage pipeline effectively identified scarce Ewing sarcoma images from other NRSTS (AUC 0.929). Compared to conventional transformer encoder architectures used for WSI representations, these SAMPLER based classifiers were three orders of magnitude faster to train, despite operating entirely without a GPU. This study highlights that digital histopathology paired with rigorous image harmonization provides a powerful solution for pediatric sarcoma classification.

Year:

2026

Resistance of BRAF-mutant melanomas to targeted therapy arises from the ability of cells to enter a persister state, evade treatment with relative dormancy, and repopulate the tumor when reactivated. A better understanding of the temporal dynamics and specific pathways leading into and out of the persister state is needed to identify strategies to prevent treatment failure. Using spatial transcriptomics in patient-derived xenograft models, we captured clonal lineage evolution during treatment. The persister state showed increased oxidative phosphorylation, decreased proliferation, and increased invasive capacity, with central-to-peripheral gradients. Phylogenetic tracing identified intrinsic and acquired resistance mechanisms (e.g., dual-specific phosphatases, reticulon-4, and cyclin-dependent kinase 2) and suggested specific temporal windows of potential therapeutic susceptibility. Deep learning–enabled analysis of histopathologic slides revealed morphologic features correlating with specific cell states, demonstrating that juxtaposition of transcriptomics and histologic data enabled identification of phenotypically distinct populations from using imaging data alone. In summary, this study defined state change and lineage selection during melanoma treatment with spatiotemporal resolution, elucidating how choice and timing of therapeutic agents will impact the ability to eradicate resistant clones. Significance: Tracking clonal progression during treatment uncovers conserved, global transcriptional changes and local clone–clone and spatial patterns underlying the emergence of resistance, providing insights into therapy-induced tumor evolution.

Year:

2025

Technological breakthroughs in spatial omics and artificial intelligence (AI) have the potential to transform the understanding of cancer cells and the tumor microenvironment. Here we review the role of AI in spatial omics, discussing the current state-of-the-art and further needs to decipher cancer biology from large-scale spatial tissue data. An overarching challenge is the development of interpretable spatial AI models, an activity that demands not only improved data integration but also new conceptual frameworks. We discuss emerging paradigms—in particular, data-driven spatial AI, constraint-based spatial AI, and mechanistic spatial modeling—as well as the importance of integrating AI with hypothesis-driven strategies and model systems to realize the value of cancer spatial information.

Year:

2025

Alternative splicing generates different mRNA isoforms from single genes, generating protein diversity essential for normal development and tissue function. Dysregulation of splicing is implicated in numerous diseases, including cancer, neurodegeneration, diabetes, and rare genetic disorders. Although thousands of spliced isoforms have been identified in disease-relevant contexts, the functional significance of most remains unknown, hindering our understanding of splicing-driven disease mechanisms and limiting therapeutic discovery. A major challenge lies in prioritizing biologically meaningful splicing events identified through short-read RNA sequencing. Existing splicing analysis tools typically rank target events based on splicing change magnitude or gene-level annotation, often without evaluating how resulting isoforms impact protein structure or function. This creates a critical bottleneck in translating splicing data into biological and therapeutic insights. To address this, we developed SpliceDecoder, a computational workflow that predicts how each splicing event or isoform impacts transcript productivity, protein sequence, and functional domains. Each event is assigned a functional effect score to guide evidence-based prioritization. SpliceDecoder facilitates a more informed interpretation of splicing data, reduces reliance on prior knowledge, and enables identification of events with potential biological and clinical relevance. We demonstrate its utility by validating known splicing alterations and identifying novel disease-associated isoform switches across public datasets.

Year:

2025

Purpose Pediatric sarcomas are rare and diverse, often leading to misclassification that hampers prognosis and treatment planning. We collected and harmonized histology images from multiple medical centers and developed accurate, generalizable classifiers for sarcoma subtype classification using deep learning techniques on digitized histology slides. Experimental Design A pediatric sarcoma histology dataset was amassed from three medical centers as well as the Children’s Oncology Group (COG), and a computational pipeline was implemented to address batch effects and standardize processing. Input parameter combinations (e.g. tile size, multi-scale feature extraction) were tested and optimized, and multiple convolutional neural network (CNN) and vision transformer-based (ViT) deep learning backbones were used as feature extractors for our SAMPLER whole slide image (WSI) representations, which we benchmarked to create an optimized, sarcoma-specific workflow. Models were trained for automated classification of rhabdomyosarcoma (RMS) versus non-rhabdomyosarcoma (NRSTS) and to further distinguish subtypes including alveolar, embryonal and spindle cell RMS as well as Ewing sarcoma. Results A total of 867 whole slide images were harmonized to generate a digital histology library with representation from 10 pediatric sarcoma subtypes. Across multiple classification tasks, ViT are superior to CNN models and multiscale feature sets consistently outperform single-scale models. Optimized classifiers accurately predicted RMS versus NRSTS (AUC 0.969±0.026) as well as alveolar versus embryonal RMS (AUC 0.961±0.021). Our two-stage classifier distinguished Ewing sarcoma from other NRSTS (AUC 0.929). Additionally, our models are computationally more lightweight than standard transformer implementations (model size 0.111 v. 1.9MB, training time > 3 orders of magnitude faster). Conclusions Digital histopathology can successfully classify pediatric sarcomas, providing results that are reproducible, mitigate inter-observer bias, and can be implemented remotely. We demonstrate that image harmonization overcomes the pitfall of overfitting to single institutional data while maintaining state-of-the-art performance across multiple classification tasks. The developed classifiers provide a basis for more precise histology-based sarcoma diagnostics, enabling global access to improved prognostication and treatment planning. Abstract Pediatric sarcomas present diagnostic challenges due to their rarity and diverse subtypes, often requiring specialized pathology expertise and costly genetic tests. To overcome these barriers, we developed a computational pipeline leveraging deep learning methods to accurately classify pediatric sarcoma subtypes from digitized histology slides. To ensure classifier generalizability and minimize center-specific artifacts, we collected and harmonized a dataset comprising 867 whole slide images (WSIs) from three medical centers and the Children’s Oncology Group (COG). Multiple convolutional neural network (CNN) and vision transformer (ViT) architectures were systematically evaluated as feature extractors for SAMPLER-based WSI representations, and input parameters such as tile size combinations and resolutions were tested and optimized. Our analysis showed that advanced ViT foundation models (UNI, CONCH) significantly outperformed earlier approaches, and incorporating multiscale features can enhance classification accuracy. Our optimized models achieved high performance, distinguishing rhabdomyosarcoma (RMS) from non-rhabdomyosarcoma (NRSTS) with an AUC of 0.969±0.026 and differentiating RMS subtypes (alveolar vs. embryonal) with an AUC of 0.961±0.021. Additionally, a two-stage pipeline effectively identified scarce Ewing sarcoma images from other NRSTS (AUC 0.929). Compared to conventional transformer encoder architectures used for WSI representations, our SAMPLER based classifiers were more lightweight (0.111 MB vs. 1.9 MB) and thr

Year:

2025

Cellular senescence is a stress response mechanism marked by irreversible growth arrest, upregulation of antiapoptotic pathways, loss of cellular function, and remodelling of the cellular secretory profile. In both humans and mice, pancreatic β-cells undergo senescence with age and insulin resistance. Targeted removal of senescent cells in mouse models of diabetes improves glucose homeostasis, demonstrating the role β-cell senescence in diabetes progression. In contrast, β-cell senescence also promotes immune surveillance, promoting β-cell survival and function. Thus, a better understanding of senescent cells’ phenotypic and functional heterogeneity is needed to develop effective therapeutic strategies. Herein, we show that subpopulations of senescent β-cells in mice and humans, which were identified through the expression of Cdkn1a (encoding p21 Cip1 ) and Cdkn2a (encoding p16 Ink4a ) by single-cell RNA sequencing (scRNA-seq), flow cytometry, spatial transcriptomics, and spatial proteomics, exhibit distinct transcriptional and functional identities. The predominant senescent β-cell subpopulation expressed Cdkn1a and was characterized by a lack of glucose responsiveness, high basal insulin secretion, and transcription of canonical SASP factors. The SASP of Cdkn1a -expressing β-cells had non-cell autonomous effects on neighbouring cells. A subset of four SASP factors from Cdkn1a + cells was sufficient to induce secondary senescence and β-cell dysfunction in vitro . JAK inhibitors (JAK1/2 and JAK1/3) counteracted secondary senescence induction and restored β-cell function in high-fat diet-fed mice and human islets from donors with or without type 2 diabetes. GRAPHICAL ABSTRACT

Year:

2025

In tumors, reciprocal spatial interactions between immune cells, their mediators, the extracellular matrix, and mutated neoplastic cells impact all aspects of treatment resistance. The operational mechanisms of these interactions are foundational for developing insights and targets for cancer therapy and prevention. Spatial quantification of the tumor microenvironment system from image data has untapped potential for patient stratification. Here, we present SparTile, a powerful computational approach for analysis of multiplex proteomics images to reveal clinically relevant structural organization in the tumor microenvironment. SparTile enables the robust unbiased identification and characterization of tumor microenvironments based on spatial relationships among protein markers without the need for cell segmentation or classification. Applied to tissues of patients with triple negative breast cancer (TNBC), an aggressive subtype of breast cancer, SparTile identified repeatable microenvironments with specific cellular relationships. These included microenvironments characterized by risk markers such as Ki67+ (p-value= 0.052) and vimentin+ (p-value<0.01) tumor cells, which correlate with poor survival. Furthermore, myeloid markers in an MX1 positive tumor environment correlate with shorter survival (p-value=0.04). Finally, we identified a relative distance between tumor and myeloid cell markers as a strong prognostic risk factor in multivariate Cox models (p-value=0.01). This distance metric was externally validated on two datasets of breast cancer multiplex images (p-values<0.01). Thus, unbiased protein-based and segmentation-free spatial analysis is effective for identifying clinically relevant biomarkers from multiplex tumor images and identifies novel predictive biology.

Year:

2025

Resistance of BRAF -mutant melanomas to targeted therapy arises from the ability of cells to enter a persister state, evade treatment with relative dormancy, and repopulate the tumor when reactivated. A better understanding of the temporal dynamics and specific pathways leading into and out of the persister state is needed to identify strategies to prevent treatment failure. Using spatial transcriptomics in patient-derived xenograft models, we captured clonal lineage evolution during treatment. The persister state showed increased oxidative phosphorylation, decreased proliferation, and increased invasive capacity, with central-to-peripheral gradients. Phylogenetic tracing identified intrinsic and acquired resistance mechanisms (e.g., dual-specific phosphatases, reticulon-4, and cyclin-dependent kinase 2) and suggested specific temporal windows of potential therapeutic susceptibility. Deep learning–enabled analysis of histopathologic slides revealed morphologic features correlating with specific cell states, demonstrating that juxtaposition of transcriptomics and histologic data enabled identification of phenotypically distinct populations from using imaging data alone. In summary, this study defined state change and lineage selection during melanoma treatment with spatiotemporal resolution, elucidating how choice and timing of therapeutic agents will impact the ability to eradicate resistant clones. Significance: Tracking clonal progression during treatment uncovers conserved, global transcriptional changes and local clone–clone and spatial patterns underlying the emergence of resistance, providing insights into therapy-induced tumor evolution.

Year:

2025

Differentially expressed genes between time point 1 and time point 0 with significant corrected p-values in the resistant lineages (1, 2) but not in the sensitive lineages (0, 5, 8) , in the sensitive but not resistant, and in both sensitive and resistant (WM4007).

Year:

2025

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2023

Year:

2023

Collaborators (40)

Lawrence Rubinstein

Statistician

National Cancer Institute

UNITED STATES
View details

Matthew H Bailey

Assistant Professor

Brigham Young University

UNITED STATES
View details

Jose G. Trevino

Virginia Commonwealth University

UNITED STATES
View details

Katherine Chiu

University of California, Davis

UNITED STATES
View details

Minji Kim

Assistant Professor

University of Michigan

UNITED STATES
View details

Michael A. Davies

-

UNITED STATES
View details

Sherri Davies

Senior Scientist

Washington University in St. Louis

UNITED STATES
View details

Min Jin Ha

Assistant Professor

The University of Texas MD Anderson Cancer Center

UNITED STATES
View details

Lisa McShane

Associate Director, Division of Cancer Treatment & Diagnosis

National Cancer Institute

UNITED STATES
View details

Peter Robinson

Professor for Medical Computer Science

The Jackson Laboratory for Genomic Medicine

UNITED STATES
View details

Emily Christison-Lagay

Yale University

UNITED STATES
View details

Xiaowei Xu

Professor

University of Pennsylvania

UNITED STATES
View details

Shidan Wang

-

UNITED STATES
View details

Paul Robson

Professor

Jackson Laboratory - Farmington

UNITED STATES
View details

Junya Fujimoto

Professor and Chair

Hiroshima University Hospital

JAPAN
View details

Alana Welm

Chair

Huntsman Cancer Institute

UNITED STATES
View details

Nicholas Mitsiades

Professor

University of California, Davis

UNITED STATES
View details

Bryan Welm

Associate Professor

Huntsman Cancer Institute

UNITED STATES
View details

Meenhard Herlyn

Professor, Director

The Wistar Institute

UNITED STATES
View details

Xuelin Huang

Professor

The University of Texas MD Anderson Cancer Center

UNITED STATES
View details

Shunqiang Li

Associate Professor

Washington University in St. Louis

UNITED STATES
View details

Jing Wang

Professor

The University of Texas MD Anderson Cancer Center

UNITED STATES
View details

Ryan C. Fields

University of Rochester

UNITED STATES
View details

Luis G Carvajal‐Carmona

Professor

University of California, Davis

UNITED STATES
View details

Funda Meric-Bernstam

The University of Texas MD Anderson Cancer Center

UNITED STATES
View details

Robert Mashl

Washington University in St. Louis

UNITED STATES
View details

Chong-xian Pan

Professor

university of california davis medical center

UNITED STATES
View details

Carol Bult

Professor and Knowlton Family Chair

Jackson Laboratory

UNITED STATES
View details

Eleanor Chen

Assistant Professor

University of Washington

UNITED STATES
View details

Salma Kaochar

Assistant Professor

Baylor College of Medicine

UNITED STATES
View details

Li Ding

Associate Professor

Washington University in Saint Louis School of Medicine

UNITED STATES
View details

J. Chuck Harrell

Virginia Commonwealth University

UNITED STATES
View details

Michael Lewis

Baylor College of Medicine

UNITED STATES
View details

George Kuchel

Professor and Director

University of Connecticut Health Center

UNITED STATES
View details

Olga Anczukow

Professor

Jackson Laboratory - Farmington

UNITED STATES
View details

Yijun Ruan

Director and Professor

Jackson Laboratory

UNITED STATES
View details

Maria Gabriela Raso

Assistant Professor

The University of Texas MD Anderson Cancer Center

UNITED STATES
View details

Soner Koc

GlaxoSmithKline Research and Development

UNITED STATES
View details

Guanghua Xiao

-

UNITED STATES
View details

Yang Xie

Professor

University of Texas Southwestern Medical Center

UNITED STATES
View details