Back

Discrete Diffusion for Single-Cell Gene Expression Modeling

Bhattacharya, S.; Gensbigler, C.; Karim, S.; Lees, J.

2026-02-20 cell biology
10.64898/2026.02.19.705033 bioRxiv
Show abstract

AO_SCPLOWBSTRACTC_SCPLOWCurrent generative modeling of single-cell transcriptomics relies on continuous latent representations, transforming inherently discrete and sparse gene counts into continuous space. We propose Discrete Cell Models (DCM), a diffusion-based framework that learns cellular representations directly in the discrete domain. Our framework supports both unconditional and conditional generation, allowing for precise modeling of complex biological scenarios such as cell-type-specific transcriptional responses to genetic perturbations. We demonstrate that DCM scales effectively and achieves strong performance against current state-of-the-art methods, including scVI, CPA, STATE, scGPT, and scLDM. On the Dentate Gyrus benchmark, DCM achieves a 5-fold improvement in MMD2RBF and a nearly 2-fold improvement in W2 distance, over the leading continuous diffusion baseline (scLDM). On the conditional Replogle perturbation benchmark, DCM sets a new state of the art on W2 distance while remaining competitive on MMD2RBF. Together, these results establish discrete diffusion as a promising direction for foundational models of cellular biology.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.