Research / Publications
Methods, applications, and perspectives in computational biology.
* Equal contribution.
STAR Protocols 2026
We present a step-by-step tMCTA-seq workflow for detecting and distinguishing colorectal and gastric cancers from plasma cell-free DNA. The protocol covers sample preparation, targeted methylation sequencing, and computational analysis, with guidance for adapting the workflow to other cancer types.
Genes 2026
We introduce stHGCL, a framework that combines a spot–gene heterogeneous graph with spatial graph learning and contrastive learning to identify spatial domains. It supports both individual tissue slices and joint analysis of multiple slices, including datasets without prior spatial alignment.
Nature Cell Biology 2026
We introduce SpaHDmap, an interpretable dimension reduction framework that enhances spatial resolution by integrating ST gene expression with high-resolution histology images. SpaHDmap incorporates non-negative matrix factorization into a multimodal fusion encoder-decoder architecture, enabling the identification of interpretable, high-resolution embeddings.
iScience 2025
We developed tMCTA-seq, a cost-effective targeted DNA methylation method for simultaneous detection and differentiation of colorectal and gastric cancers. By integrating a universal nested primer design with an ensemble machine learning framework, we achieved high diagnostic performance across multiple cancer stages in clinical cfDNA samples.
Briefings in Bioinformatics 2024
By establishing a heterogeneous graph that represents the interactions between multiple batches of cells and genes, and combining a heterogeneous graph neural network with contrastive learning, we proposed a structure-preserved scRNA-seq data integration approach scHetG.
Clinical Cancer Bulletin 2024
We review spatial transcriptomics technologies and their applications to tumor heterogeneity and the tumor microenvironment. The article also outlines computational analysis workflows, provides an accompanying analysis tutorial, and discusses opportunities for spatial multi-omics in cancer research.
BMC Bioinformatics 2024
To expedite the exploration of systematic disparities under various biological contexts, we propose a scRNA-seq integration method called scDisco, which involves a domain-adaptive decoupling representation learning strategy for the integration of dissimilar scRNA-seq data.
NAR Genomics and Bioinformatics 2024
We introduce scBiG, a novel graph node embedding method designed for representation learning in scRNA-seq data. scBiG establishes a bipartite graph connecting cells and expressed genes, and then constructs a multilayer graph convolutional network to learn cell and gene embeddings.
Bioinformatics 2023
We propose an adaptive embedding and clustering method named scAce, which constructs a variational autoencoder to simultaneously learn cell embeddings and cluster assignments. In the scAce method, we develop an adaptive cluster merging approach which achieves improved clustering results without the need to estimate the number of clusters in advance.
Genome Biology 2022
Although different batch effect removal methods have been developed, none are suitable for heterogeneous single-cell samples coming from multiple biological conditions. We propose a method, scINSIGHT, to learn coordinated gene expression patterns that are common among, or specific to, different biological conditions, and identify cellular identities and processes across single-cell samples.