scDisentangle: sequential disentanglement for single-cell perturbation response prediction
- 1University Mohammed VI Polytechnic
Abstract
Predicting the transcriptional response of single cells to perturbation is a central goal of single-cell modeling, with broad application to analysis and experimental design. Existing methods often fail to beat simple baselines and offer little insight into how biological variation is organized. We present scDisentangle, a model that decomposes single-cell expression into observed attributes and a sequence of residual components, generating counterfactuals that preserve source cell identity. scDisentangle outperformed current models and naive baselines on cytokine stimulation, liver infection, and unseen combinatorial genetic perturbations, on which accuracy was largely independent of how closely a response resembled the average perturbation effect. In an immune checkpoint inhibitor-associated myocarditis cohort, cell-resolved counterfactuals predicted donor-specific intracardiac CD8+ T cell states from peripheral blood. In a castration-regeneration prostate atlas, predictions recapitulated luminal dynamics and progenitor-marker kinetics upon androgen add-back. scDisentangle's coarse-to-fine latent structure aligned with hematopoietic lineage organization, with early levels capturing a less differentiated core progressively refined into cell-type-specific programs.
Related
- The recoverable resolution of cellular perturbation-response prediction
- A biological-response compound representation allows chemical perturbation prediction across cell lines
- Generalizable single-cell perturbation response prediction using energy-guided flow matching
- ISP Platform powered by Geneformer: Framework for Cross-Species, Sequential, and Multi-Gene In Silico Perturbation Screens with Application to iPS Cell State Transitions
- PerturbBridge: Conditional Latent Schr\"{o}dinger Bridge for Single-Cell Perturbation Prediction
BibTeX
@misc{mejjad2026scdisentangle,
title = {{scDisentangle: sequential disentanglement for single-cell perturbation response prediction}},
author = {S. El Mejjad and S. Enezari and A. Bedraoui and F. Z. El Hajji and R. El Fatimy and M. Hofree and T. Daouda},
howpublished = {bioRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.04.764292},
url = {https://www.biorxiv.org/content/10.64898/2026.11.04.764292v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - scDisentangle: sequential disentanglement for single-cell perturbation response prediction
AU - S. El Mejjad
AU - S. Enezari
AU - A. Bedraoui
AU - F. Z. El Hajji
AU - R. El Fatimy
AU - M. Hofree
AU - T. Daouda
PY - 2026
DA - 2026/11/04
PB - bioRxiv
DO - 10.64898/2026.11.04.764292
UR - https://www.biorxiv.org/content/10.64898/2026.11.04.764292v1
AB - Predicting the transcriptional response of single cells to perturbation is a central goal of single-cell modeling, with broad application to analysis and experimental design. Existing methods often fail to beat simple baselines and offer little insight into how biological variation is organized. We present scDisentangle, a model that decomposes single-cell expression into observed attributes and a sequence of residual components, generating counterfactuals that preserve source cell identity. scDisentangle outperformed current models and naive baselines on cytokine stimulation, liver infection, and unseen combinatorial genetic perturbations, on which accuracy was largely independent of how closely a response resembled the average perturbation effect. In an immune checkpoint inhibitor-associated myocarditis cohort, cell-resolved counterfactuals predicted donor-specific intracardiac CD8+ T cell states from peripheral blood. In a castration-regeneration prostate atlas, predictions recapitulated luminal dynamics and progenitor-marker kinetics upon androgen add-back. scDisentangle's coarse-to-fine latent structure aligned with hematopoietic lineage organization, with early levels capturing a less differentiated core progressively refined into cell-type-specific programs.
ER -