Luni Hu

DCAF: Unifying Data Validity, Evaluation Validity and Predictive Modeling for Single-cell Perturbation Simulation

DCAF improves single-cell perturbation simulation by addressing data, evaluation, and model validity, outperforming baselines in generalization and recapitulating regulatory heterogeneity in mouse brain.

RegFormer: A Single-Cell Foundation Model Powered by Gene Regulatory Hierarchies

RegFormer improves single-cell RNA sequencing analysis by integrating gene regulatory networks with Mamba-based modeling, outperforming existing models in clustering, integration, and annotation.

BiOmics: A Foundational Agent for Grounded and Autonomous Multi-omics Interpretation

BiOmics introduces an AI agent that improves multi-omics interpretation by combining reasoning and embedding spaces, outperforming existing methods in variant discovery, cell annotation, and drug repurposing.

DynPerturb:Dynamic Perturbation Modeling for Spatiotemporal Single-Cell Systems

DynPerturb models spatiotemporal perturbation dynamics in single-cell and spatial transcriptomics, outperforming existing methods in regulatory link prediction and uncovering dynamic regulatory rewiring in biological systems.

BioLLM: A Standardized Framework for Integrating and Benchmarking Single-Cell Foundation Models

BioLLM integrates and benchmarks single-cell foundation models, revealing scGPTs' robust performance and highlighting limitations of scBERT in gene-level tasks.

UniCell: Towards a Unified Solution for Cell Annotation, Nomenclature Harmonization, Atlas Construction in Single-Cell Transcriptomics

UniCell improves cell type annotation and cross-dataset harmonization by combining Cell Ontology with transcriptomic data, outperforming existing tools in accuracy and rare cell detection.

Stereopy: modeling comparative and spatiotemporal cellular heterogeneity via multi-sample spatial transcriptomics

Stereopy provides a framework for analyzing multi-sample spatial transcriptomics data to model comparative and spatiotemporal cellular heterogeneity through three key algorithms.

ST-GEARS: Advancing 3D Downstream Research through Accurate Spatial Information Recovery

ST-GEARS improves spatial information recovery in 3D spatial transcriptomics by aligning sections and correcting distortions, enhancing accuracy for downstream analysis.