Mingze Dong

Yale University.

headshot0110.png

I received my Ph.D. in Computational Biology from Yale University, where I was co-advised by Prof. Yuval Kluger and Prof. Rong Fan. Before Yale, I earned my B.S. from Peking University, focusing on applied mathematics and biology.

I am currently a postdoctoral fellow at Stanford University before launching into independence. I am also building foundation models for biology at Arc.

I believe AI has enormous potential to transform biology, but the field has yet to achieve a true GPT moment. I think the gap is not just scale: it reflects limited biological data, the direct adoption of model architectures from other fields, and insufficient attention to what models actually learn — real biology versus batch effects, or generalization versus memorization.

My work approaches this question through large-scale AI, data science, and computational biology, with the goal of collecting the right data and building state-of-the-art models that help us understand, predict, and eventually design biological systems.

I am very happy to connect with people who share this vision and are excited about the opportunities ahead.

selected publications

  1. Preprint
    Stack: In-context learning of single-cell biology
    Mingze Dong, Abhinav Adduri, Dhruv Gautam, and 6 more authors
    Preprint, 2026
  2. Nature Methods
    Causal identification of single-cell experimental perturbation effects with CINEMA-OT
    Mingze Dong, Bao Wang, Jessica Wei, and 8 more authors
    Nature Methods, 2023
  3. Preprint
    Scaling deep identifiable models enables zero-shot characterization of single-cell biological states
    Mingze Dong, Kriti Agrawal, Rong Fan, and 3 more authors
    bioRxiv, 2024
  4. Nature Commun
    Editor’s Highlight
    SIMVI disentangles intrinsic and spatial-induced cellular states in spatial omics data
    Mingze Dong, David Su, Harriet Kluger, and 2 more authors
    Nature Communications (Featured Article, <5%), 2025
  5. Nature Methods
    Pertpy: an end-to-end framework for perturbation analysis
    Lukas Heumos, Yuge Ji, Lilly May, and 25 more authors
    Nature Methods, 2025
  6. NeurIPS
    Understanding and enhancing mask-based pretraining towards universal representations
    Mingze Dong, Leda Wang, and Yuval Kluger
    The Thirty-Ninth Annual Conference on Neural Information Processing Systems, 2025
  7. ICML
    Towards Understanding and Reducing Graph Structural Noise for GNNs
    Mingze Dong, and Yuval Kluger
    Proceedings of the 40th International Conference on Machine Learning, 2023