CancerZigZag: Iterative Seed-Anchored Diffusion for Generative Modeling of Single-Cell State Transitions
Abstract
Single-cell cancer datasets are predominantly cross-sectional and rarely provide paired or longitudinal observations linking individual healthy-like cells to tumor-associated states. We introduce CancerZigZag, a seed-initialized diffusion-based framework for exploratory generation of tumor-associated single-cell candidate clouds from unpaired epithelial cell populations. For each cancer context, a diffusion model is trained exclusively on tumor-derived epithelial cells and applied through repeated partial latent-space perturbation and reverse diffusion to held-out healthy-like seeds, generating stochastic candidate clouds without paired measurements or classifier guidance. We applied CancerZigZag to colorectal, breast, lung, and renal cell carcinoma contexts and explored parameter landscapes defined by perturbation depth and the number of ZigZag cycles. Across the reported operating configurations, candidate clouds contained outputs classified toward held-out tumor-derived reference populations for each evaluated seed. Residual seed-dependent organization varied across contexts, with the clearest structure in colorectal cancer, more modest organization in lung cancer, and limited cloud-level structure in breast and renal cell carcinoma. Representative candidates also showed directional concordance with transcriptional shifts observed between held-out healthy-like and tumor-derived reference populations. CancerZigZag is not interpreted as a model of deterministic healthy-to-tumor transformation or cellular progression. Instead, it provides a reference-informed framework for exploring tumor-associated candidate distributions from unpaired healthy-like seeds and quantifying the context-dependent relationship between tumor-associated displacement and residual seed dependence.
Related
- Protein Deimmunization with a Generative Diffusion Model
- Differentiable Spin-Center Diffusion Model Enables DEER-Guided Generative Pipeline of Weighted Protein Ensembles
- ScaffoldForge: Bioinspired generative design of 3D-printed bone scaffolds via conditional diffusion for large segmental bone defect reconstruction
- Clinical-Anchored Incremental Survival Learning: A Stability-Constrained Framework for Molecularly Enhanced Survival Prediction in Lung Adenocarcinoma
- Genomic imprinting of the metabolic regulator gene Klf14 is regulated by a paternal sub-TAD anchored at Mest
BibTeX
@misc{schluter2026cancerzigzag,
title = {{CancerZigZag: Iterative Seed-Anchored Diffusion for Generative Modeling of Single-Cell State Transitions}},
author = {Johannes Schlüter and Alexander Schönhuth},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.37735},
eprint = {2609.37735},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.37735},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - CancerZigZag: Iterative Seed-Anchored Diffusion for Generative Modeling of Single-Cell State Transitions
AU - Johannes Schlüter
AU - Alexander Schönhuth
PY - 2026
DA - 2026/09/29
PB - arXiv
DO - 10.48550/arxiv.2609.37735
UR - https://arxiv.org/abs/2609.37735
AB - Single-cell cancer datasets are predominantly cross-sectional and rarely provide paired or longitudinal observations linking individual healthy-like cells to tumor-associated states. We introduce CancerZigZag, a seed-initialized diffusion-based framework for exploratory generation of tumor-associated single-cell candidate clouds from unpaired epithelial cell populations. For each cancer context, a diffusion model is trained exclusively on tumor-derived epithelial cells and applied through repeated partial latent-space perturbation and reverse diffusion to held-out healthy-like seeds, generating stochastic candidate clouds without paired measurements or classifier guidance. We applied CancerZigZag to colorectal, breast, lung, and renal cell carcinoma contexts and explored parameter landscapes defined by perturbation depth and the number of ZigZag cycles. Across the reported operating configurations, candidate clouds contained outputs classified toward held-out tumor-derived reference populations for each evaluated seed. Residual seed-dependent organization varied across contexts, with the clearest structure in colorectal cancer, more modest organization in lung cancer, and limited cloud-level structure in breast and renal cell carcinoma. Representative candidates also showed directional concordance with transcriptional shifts observed between held-out healthy-like and tumor-derived reference populations. CancerZigZag is not interpreted as a model of deterministic healthy-to-tumor transformation or cellular progression. Instead, it provides a reference-informed framework for exploring tumor-associated candidate distributions from unpaired healthy-like seeds and quantifying the context-dependent relationship between tumor-associated displacement and residual seed dependence.
ER -