Ruochi Zhang

Bolero predicts cell-state-specific gene regulation from DNA across species by integrating cell-state embeddings and biological variables.

The paper constructs a single-nucleus multi-omic atlas of adult human tissues, identifying regulatory elements and variant effects across 21 tissues.

This paper introduces a computational method to predict higher-order genetic interactions, demonstrating its effectiveness in yeast by predicting trigenic interactions and uncovering biological functions related to cell growth.

The paper identifies transcription factor collaborations that shape CD8+ T cell states and demonstrates how engineering these interactions can influence cell fate for therapeutic applications.

This paper introduces DiffEvol, a constrained diffusion framework to model evolution, and applies it to SARS-CoV-2 data to reconstruct fitness trends and vaccine-induced phase transitions.

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.

This study identifies engineered histone variants that modulate chromatin structure and function, revealing sequence-to-function patterns and enabling the design of synthetic histones for specific chromatin states.

The paper shows that colitis leaves an epigenetic memory in colonic stem cells that promotes tumor growth by enhancing regenerative gene programs and AP-1 activity.

The paper introduces ExIGS, a method linking nuclear abnormalities to chromatin repression hotspots in progeria fibroblasts.

SCGHOST identifies single-cell 3D genome subcompartments using scHi-C data and provides insights into cell-to-cell variability in nuclear organization.

The paper develops an antigen-specific assay for T cell exhaustion and identifies IKAROS as a driver of exhaustion, showing that iberdomide prevents exhaustion by blocking chromatin remodeling.

PhyGCN improves node representations in hypergraphs using self-supervised learning and variable hyperedge sizes, showing effectiveness on chromatin interactions and polypharmacy side-effects.

This paper introduces GAGE-seq, a single-cell method that simultaneously profiles 3D genome structure and gene expression, revealing their relationship in defining cell identity and lineage commitment.

The paper introduces PRINT, a deep learning method to correct chromatin accessibility data and analyze DNA-protein interactions, revealing structural and functional changes in cis-regulatory elements during hematopoiesis and aging.

Fast-Higashi improves scHi-C analysis by jointly identifying cell identities and chromatin meta-interactions with tensor decomposition and a novel algorithm, enhancing rare cell type detection and developmental trajectory reconstruction.

Higashi improves scHi-C data analysis by enhancing quality and extracting 3D genome features, revealing cell-to-cell variability and new gene regulation connections in human prefrontal cortex cells.

The paper introduces SPIN, a method to identify genome-wide nuclear spatial localization patterns by integrating nuclear compartment mapping and chromatin interaction data.

The paper introduces MATCHA, a method for analyzing multi-way chromatin interactions using hypergraph representation learning, improving data denoising and prediction accuracy.

MOCHI identifies heterogeneous interactome modules in the nucleus by clustering gene loci with frequent spatial contacts and shared transcription factor regulation across five cell types.

This study identifies sequence-based features that predict CTCF-mediated chromatin loops using a machine learning model called CTCF-MP.