Xiuwei Zhang

CeSpGRN infers cell-specific gene regulatory networks from single-cell multi-omics and spatial data, showing improved accuracy over existing methods.

PantheonOS introduces a multi-agent framework for automated genomics analysis, achieving super-human performance in batch correction and gene panel selection across diverse biological tasks.

LineageMap infers spatially resolved cell lineage trees from tri-modal single-cell data, outperforming existing methods in accuracy and revealing biologically coherent trajectories.

SpaDecoder improves cell type deconvolution in spatial transcriptomics by leveraging 3D tissue structure and single-cell reference profiles.

EpiPack improves scATAC-seq analysis by integrating heterogeneous features and detecting perturbed cell types and states through transfer learning and graph-based modeling.

TemSOMap integrates lineage data with spatial transcriptomics to improve cell spatial coordinate inference.

scDisInFact disentangles batch and condition effects in scRNA-seq data, improving batch correction, gene detection, and perturbation prediction across multiple batches and conditions.

LinRace improves lineage reconstruction by integrating lineage barcodes and gene expression data, producing more accurate cell division trees and inferring ancestral cell states.

scMultiSim simulates multi-modal single-cell data by integrating cell-cell interactions and gene regulatory networks, enabling broader benchmarking of computational methods.

The paper introduces scDART, a deep learning framework that integrates scRNA-seq and scATAC-seq data and learns cross-modal relationships for trajectory inference.

scMoMaT integrates single-cell multi-omics data under mosaic integration using matrix tri-factorization, uncovering modality-specific biomarkers for improved cell type annotation.

TedSim simulates cell division events and generates scRNA-seq and lineage barcode data to benchmark methods for cell lineage analysis.

CellPath infers high-resolution cell developmental trajectories using RNA velocity, allowing multiple paths and automatic root detection without topology constraints.

VeloSim simulates gene-expression and RNA velocity along trajectories to benchmark computational methods with controlled biological and technical variation.

This paper introduces scHybridNMF, a method that combines gene expression and spatial location data to improve cell type clustering in single-cell data.

The paper presents an Integer Linear Programming approach for optimal ancestral PPI network reconstruction under the DMC model, showing improved likelihood and agreement with biological evidence.

GRNUlar infers gene regulatory networks from scRNA-Seq data using unrolled deep learning, outperforming existing methods with and without TF information.

SymSim simulates single cell RNA-Seq data with controlled variability to benchmark clustering and differential expression methods and assess experimental design parameters.