Jin Gu

This study evaluates how large language models generate summaries of bioinformatics analyses, finding they produce readable and safe summaries but often misinterpret visualizations and lack novel insights.

This paper identifies a drug combination that irreversibly senesces drug-tolerant persister cells, preventing tumor relapse in melanoma models.

SpatialZoomer identifies disease progression signals and spatially heterogeneous niches in spatial transcriptomics data by analyzing features at multiple scales.

scTrace+ improves cell fate inference by integrating lineage-tracing data with transcriptomic similarities using a Kernelized Probabilistic Matrix Factorization model, accurately predicting cell fates and identifying driver genes.

GeST is a generative transformer model pretrained on spatial transcriptomics data to learn spatial context, showing improved generation and zero-shot spatial niche embedding extraction.

uniLIVER creates a human liver cell atlas using sc/snRNA-seq data and develops LiverCT to map abnormal datasets to the reference atlas, revealing changes in cell type composition and states in liver cancer.

scCancer2 provides a supervised framework for annotating tumor microenvironment cells across datasets and evaluates subtype relationships using scRNA-seq and bulk data.

This paper introduces HoloNet, a method to decode functional cell-cell communication events from spatial transcriptomic data by integrating ligand-receptor pairs, cell-type spatial distribution, and downstream gene expression.

TIST integrates spatial transcriptomics data with histopathological images to improve spatial cluster identification and gene expression analysis.

The paper introduces JEBIN, a method to integrate multiple gene expression datasets into a low-dimensional gene representation, showing improved performance in co-expression network analysis and differential co-expression studies.

ARIC estimates cell type proportions from bulk gene expression or DNA methylation data, showing accuracy in both data types.

CITE-sort improves cell type classification in CITE-seq by addressing artificial cell types caused by multiplets, achieving better clustering performance than existing methods.

scCancer processes and analyzes scRNA-seq data for cancer research, identifying cell populations, estimating malignancy scores, and analyzing intra-tumor heterogeneity.

SIP provides an interchangeable pipeline for scRNA-seq data processing, enabling easy integration and customization of analysis steps.

VASC is a deep variational autoencoder for scRNA-seq data that improves dimension reduction and visualization by modeling dropout events and nonlinear features.

The paper identifies recurrently deregulated lncRNAs in HCC and PVTT samples, linking them to metastasis and prognosis using multi-omics data.