Shihua Zhang

STAVelo infers RNA velocity from spatial transcriptomics data to capture spatiotemporal cellular dynamics across multiple species and tissues.

STADiffuser simulates spatial transcriptomic data with high fidelity and enables 3D modeling of large-scale spatial datasets.

This paper introduces STACAME, a framework for analyzing cross-species spatial transcriptomics data to identify shared and divergent tissue architectures across species.

CellLoop identifies single-cell chromatin loops using a density-based framework, improving accuracy in mouse brain and embryogenesis data.

CAMEX integrates, aligns, and annotates multi-species scRNA-seq data using a GNN, outperforming existing methods and enabling species-specific cell type detection.

STAX is a multi-task learning platform for spatial omics that addresses multiple analytical challenges, offering superior performance and biological insights.

This paper evaluates single-cell foundation models and finds they struggle to remove batch effects, limiting their ability to create universal cellular embeddings.

STAMapper maps and annotates cell types in single-cell spatial transcriptomics data using a graph neural network, achieving high accuracy and enabling unknown cell-type detection.

CAMUS improves cross-dataset cell type annotation by selecting optimal references and methods, achieving accuracy gains across multiple single-cell datasets.

This paper introduces STAVAG, a framework to identify directionally and temporally variable genes in spatial and temporal data, revealing biologically relevant gene modules and diagnostic biomarkers.

STALocator localizes single cells to spatial transcriptomics data, improving spatial reconstruction and gene expression analysis in multiple datasets.

TADGATE identifies TADs from sparse Hi-C maps, enhances their structure, and reveals functional elements and cell-type-specific regulatory mechanisms.

scCompass provides a cross-species scRNA-seq database with standardized preprocessing, enabling AI-ready data for research and analysis.

This study shows that RNF220 regulates WDR5 stability to maintain Hox gene expression patterns in the mouse hindbrain, with implications for pons development.

STASCAN improves spatial transcriptomics resolution by combining gene expression and histology images, revealing finer cell structures across species and tissues.

The paper introduces STADIA, a method for integrating spatial transcriptomics data across batches and slices, improving domain detection and reducing batch effects.

The paper introduces STMask, a graph self-supervised learning method for spatial transcriptomics data, improving spatial clustering and trajectory inference.

This paper develops STAligner, a graph attention neural network for integrating spatial transcriptomics data across different conditions, technologies, and developmental stages, identifying shared and specific spatial domains.

CellPolaris uses transfer learning to generate cell-type-specific gene regulatory networks from transcriptomic data, achieving state-of-the-art results in predicting regulatory factors and simulating cell fate transitions.

STAMarker identifies spatial domain-specific variable genes using saliency maps in deep learning, outperforming existing methods on sparse spatial transcriptomic data.

SmartGate improves spatial segmentation and biomarker identification in IMS data through iterative peak picking and graph attention auto-encoder modeling.

The paper defines a TAD separation landscape to address inconsistencies in TAD identification across methods and identifies Consensus Topological Associating Domains.

The paper develops HiCS, a method to identify hierarchical chromatin domains in single-cell Hi-C maps, revealing multiple factors and retrotransposons involved in boundary formation in mouse embryonic stem cells.

This study identifies STAT3 as a positive regulator of RNA polymerase III transcription through the miR-106a-5p/TP73 axis, showing STAT3 inhibits TP73 and miR-106a-5p activates Pol III transcription.

This paper introduces CAME, a heterogeneous graph neural network for cross-species cell-type assignment in scRNA-seq data, showing improved performance and revealing evolutionary patterns.

This paper introduces STAGATE, a graph attention auto-encoder for spatial domain detection in spatial transcriptomics, improving accuracy and reducing batch effects across sections.

VP-Detector uses a 3D convolutional neural network to automatically localize and classify macromolecules in cryo-electron tomograms, achieving competitive performance with efficient processing.

The paper introduces scAND, a network diffusion method for embedding single-cell ATAC-seq data, improving clustering accuracy and handling sparsity and missingness.

PBLR imputes dropout events in scRNA-seq data by considering cell heterogeneity and prior expression levels, improving low-dimensional representation and subpopulation detection.

This paper jointly reconstructs cis-regulatory interaction networks across multiple tissues using single-cell ATAC-seq data, revealing common and tissue-specific regulatory mechanisms.