Mingze Dong

Predicting cellular responses to perturbation across diverse contexts with STATE

State, a set-based transformer trained on over 100 million perturbed cells, predicts perturbation responses in new cellular contexts better than mean, linear and deep learning baselines.

Stack: In-Context Learning of Single-Cell Biology

SO_SCPLOWTACKC_SCPLOW is a foundation model trained on human single-cell data that enables in-context learning for predicting perturbation effects without fine-tuning.

Human Lymph Node Cellular Senescence Atlas Reveals Age-Dependent Alteration in Germinal Center B Cell Function and Niches

This study maps age-related cellular senescence in human lymph nodes, identifying senescence markers and spatial shifts in germinal center B cells.

Cross-Species Multi-Omics Profiling Identifies Conserved Activated Valvular Interstitial Cell Population Driving Myxomatous Mitral Valve Degeneration

The paper identifies a conserved activated valvular interstitial cell population across species and disease types that drives myxomatous mitral valve degeneration.

Pertpy: an end-to-end framework for perturbation analysis

Pertpy provides a modular framework for analyzing large-scale perturbation single-cell experiments, offering tools for metadata annotation and perturbation distance calculations.

Hormetic elevation of taurine restrains inflammaging by deactivating the NLRP3 inflammasome.

This study shows that elevated taurine reduces age-related inflammation by inhibiting the NLRP3 inflammasome and pyroptosis in mice.

Scaling deep identifiable models enables zero-shot characterization of single-cell biological states

This paper introduces scShift, a deep variational model that enables zero-shot characterization of single-cell biological states across datasets by overcoming batch effects.

Causal identification of single-cell experimental perturbation effects with CINEMA-OT

CINEMA-OT identifies causal effects of single-cell perturbations by separating confounding variation, outperforming existing methods on simulated and real datasets.

SIMVI reveals intrinsic and spatial-induced states in spatial omics data

SIMVI disentangles intrinsic and spatial-induced variations in spatial omics data, revealing spatial effects and biological dynamics across multiple datasets.