Jie Liu

Reference-Guided De Novo Retrieval of 3D Chromatin Organizational Features with Representation Learning

The paper introduces C-Fetch, a framework that retrieves 3D chromatin features using reference examples, showing improved enrichment for biologically relevant contacts compared to existing methods.

A single-cell multimodal foundation model enables cross modality analysis between transcriptome and chromatin

MUTUAL enables cross-modality analysis between transcriptome and chromatin by learning a shared RNA-ATAC space from multiome data without refitting.

Deep-Plant: a supervised foundation model for plant regulatory genomics

This paper introduces a supervised foundation model for plant regulatory genomics that predicts chromatin state from genomic sequence, showing improved speed, accuracy, and interpretability over DNA language models.

An integrated view of the structure and function of the human 4D nucleome

The paper integrates 4D Nucleome data to analyze 3D genome structure and function in human cells, revealing chromatin domains, loops, and interactions, and showing how genetic variants affect genome folding.

Developing a general AI model for integrating diverse genomic modalities and comprehensive genomic knowledge

The paper develops a general AI model that integrates diverse genomic modalities and knowledge, showing strong performance across cell types and tissues.

CelLink: integrating single-cell multi-omics data with weak feature linkage and imbalanced cell populations

CelLink integrates single-cell multi-omics data by addressing weak feature linkage and imbalanced cell populations, outperforming existing methods in imputation accuracy and cell subtype annotation.

Genome structure mapping with high-resolution 3D genomics and deep learning

The paper introduces Cleopatra, a deep learning model that predicts high-resolution 3D genome maps across human cell types, revealing cell-type-specific loops and gene regulation patterns.

A generalizable framework to comprehensively predict epigenome, chromatin organization, and transcriptome

EPCOT is a deep learning framework that predicts epigenome, chromatin organization, transcriptome, and enhancer activity, showing generalizability across cell types and modalities.

NucleoMap: a computational tool for identifying nucleosomes in ultra-high resolution contact maps

NucleoMap identifies nucleosome positions from ultra-high resolution contact maps, outperforming existing methods and revealing associations between nucleosome organization and chromatin functions.

Characterizing collaborative transcription regulation with a graph-based deep learning approach

The paper introduces ECHO, a graph-based deep learning model that predicts chromatin features using 3D chromatin organization, showing improved performance over sequence-based models.

Connecting high-resolution 3D chromatin organization with epigenomics

The paper develops a deep learning model, CAESAR, to predict high-resolution 3D chromatin structures from epigenomic data and validates its accuracy across 91 human tissues and cell lines.

scHiCTools: a computational toolbox for analyzing single cell Hi-C data

scHiCTools provides a computational toolbox for analyzing single-cell Hi-C data, offering methods for smoothing, embedding, clustering, and visualization of chromatin organization.

Three-dimensional regulation of HOXA cluster genes by a cis-element in hematopoietic stem cell and leukemia.

The paper investigates how 3D chromatin interactions regulate HOXA cluster genes in HSCs and leukemia by analyzing the role of the Geneless Canyon region.

Comprehensive statistical inference of the clonal structure of cancer from multiple biopsies

This paper introduces THEMIS, a method for analyzing tumor heterogeneity across multiple biopsies, showing improved accuracy over existing methods and applications in tracking cancer progression and treatment resistance.