Hongkai Ji

Professor

Johns Hopkins University Bloomberg School of Public Health
Country flagUnited States
ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

Articles (5)

Spatially resolved assay for transposase-accessible chromatin with sequencing (spATAC-seq) is an emerging technology for studying spatial variation in gene regulatory landscapes within tissues. Current analysis pipelines commonly apply library size normalization, assuming that variation in sequencing library size across cells represents a technical confounder rather than biological signal. While recent studies have shown that library size can confound biological interpretation in spatial transcriptomics, its impact in spATAC-seq remains poorly understood. Here, we show that library size in spATAC-seq data is biologically informative and that standard normalization methods can obscure important biological signals and hinder downstream analyses. These findings underscore the need for caution and for the development of improved approaches to address library size in spATAC-seq analysis.

Year:

2025

Purpose: Co-mutations of the Kirsten rat sarcoma virus ( KRAS ) and serine / threonine kinase 11 ( STK11 ) genes in advanced non–small cell lung cancer (NSCLC) are associated with immune checkpoint blockade (ICB) resistance. Although neoadjuvant chemoimmunotherapy is now a standard-of-care treatment for resectable NSCLC, the clinical and immunologic impacts of KRAS and STK11 co-mutations in this setting are unknown. Experimental Design: We evaluated and compared recurrence-free survival of resectable KRAS -mutated NSCLC tumors, with or without co-occurring STK11 mutations, treated with neoadjuvant ICB. Single-cell transcriptomics was performed on tumor-infiltrating T cells from seven KRAS mut /STK11 wt tumors and six KRAS and STK11 co-mutated ( KRAS mut /STK11 mut ) tumors. Results: Relative to KRAS mut /STK11 wt tumors, KRAS mut /STK11 mut exhibited significantly higher recurrence risk. Single-cell transcriptomics showed enhanced oxidative phosphorylation with evidence of decreased prostaglandin E2 signaling and increased IL-2 signaling in CD8 + tumor-infiltrating lymphocytes (TIL) from KRAS mut /STK11 mut tumors, a finding that was mirrored in KRAS wt tumors that relapsed. TILs from KRAS mut /STK11 mut tumors expressed high levels of molecules associated with tumor residence, including CD39 and ZNF683 (HOBIT). Conclusions: These divergent T-cell transcriptional fates suggest that T-cell maintenance and residence may be detrimental to antitumor immunity in the context of neoadjuvant ICB for resectable NSCLC, regardless of KRAS mutation status. Our work provides a basis for future investigations into the mechanisms underpinning prostaglandin E2 signaling and IL-2 signaling as they relate to T-cell immunity to cancer and to divergent clinical outcomes in KRAS mut /STK11 mut NSCLC treated with neoadjuvant ICB.

Year:

2025

Evolutionary conservation is crucial for understanding genome functions and lays the foundation for using animal models in studying human diseases. However, conventional conservation scores based on DNA sequence evolution do not capture the dynamic biochemical activities of DNA elements, termed functional conservation. Quantifying functional conservation has been limited by the availability of functional genomic data matched across species. To address this, we developed FUNCODE, a framework for characterizing functional conservation through in silico sample matching. Applying FUNCODE to 2,595 uniformly processed datasets from the Encyclopedia of DNA Elements (ENCODE), we generated genome-wide FUNCODE scores for human and mouse regulatory elements, identifying 3.3 million functionally conserved human-mouse element pairs. We demonstrate FUNCODE’s diverse applications, including annotating 78,501 novel regulatory elements, transferring 37,968 high-resolution human ENCODE Hi-C loops in immune lineages to mice, identifying conserved functional signals for disease modeling, and enhancing cross-species integration of single-cell omics data.

Year:

2024

Background Selecting highly variable features is a crucial step in most analysis pipelines of single-cell RNA-sequencing (scRNA-seq) data. Despite numerous methods proposed in recent years, a systematic understanding of the best solution is still lacking. Results Here, we systematically evaluate 47 highly variable gene (HVG) selection methods, consisting of 21 baseline methods developed based on different data transformations and mean-variance adjustment techniques and 26 hybrid methods developed based on mixtures of baseline methods. Across 19 diverse benchmark datasets, 18 objective evaluation criteria per method, and 5,358 analysis settings, we observe that no single baseline method consistently outperforms the others across all datasets and criteria. However, hybrid methods as a group robustly outperform individual baseline methods. Based on these findings, a new HVG selection approach, mixture HVG selection (mixHVG), that incorporates top-ranked features from multiple baseline methods is proposed as a better solution to HVG selection. An open source R package mixhvg is developed to enable convenient use of mixHVG and its integration into users’ data analysis pipelines. Conclusion Our benchmark study not only provides a systematic comparison of existing methods, leading to a better HVG selection solution, but also creates a pipeline and resource consisting of diverse benchmark data and criteria for evaluating new methods in the future.

Year:

2024

Recently, RNA velocity has driven a paradigmatic change in single-cell RNA sequencing (scRNA-seq) studies, allowing the reconstruction and prediction of directed trajectories in cell differentiation and state transitions. However, most existing methods use dynamic modeling via ordinary differential equations (ODE) for individual genes in sequence and can lead to erroneous results, as they are inadequately able to fully capture the intrinsically stochastic nature of transcriptional dynamics governed by a cell-specific latent time across multiple genes. Here, we present SDEvelo, a novel deep generative approach to inferring RNA velocity by modeling the dynamics of unspliced and spliced RNAs via multivariate stochastic differential equations (SDE). Uniquely, SDEvelo explicitly models inherent uncertainty in transcriptional dynamics while estimating a cell-specific latent time across genes. Using both simulated and four scRNA-seq and spatial transcriptomics datasets, we show that SDEvelo can model the random dynamic patterns of mature-state cells while accurately detecting carcinogenesis. Additionally, the estimated gene-shared latent time can facilitate many downstream analyses for biological discovery. We demonstrate that SDEvelo is computationally scalable and applicable to both scRNA-seq and sequencing-based spatial transcriptomics data.

Year:

2024

Collaborators (10)

Joshua Reuss

Assistant Professor of Medicine

-

UNITED STATES
View details

Khaled Sanber

Assistant Professor

The University of Texas MD Anderson Cancer Center

UNITED STATES
View details

Zhicheng Ji

Assistant Professor

Duke University School of Medicine

UNITED STATES
View details

Janis M. Taube

Johns Hopkins University

UNITED STATES
View details

Chengqi Lin

Professor

Southeast University

CHINA
View details

Valsamo Anagnostou

Alex Grass Professor of Oncology

Johns Hopkins University School of Medicine

UNITED STATES
View details

Tianbei Zhang

Johns Hopkins University

UNITED STATES
View details

Patrick M. Forde

Trinity College Dublin

IRELAND
View details

Jonathan Spicer

Professor of Surgery

McGill University

CANADA
View details

Drew Pardoll

Professor

Johns Hopkins University

UNITED STATES
View details