The recoverable resolution of cellular perturbation-response prediction
- 1University of Pennsylvania Perelman School of Medicine
Abstract
Virtual cell models aim to predict how cells respond to perturbations, yet accurate prediction of post-perturbation gene expression can conceal failures to recover context-specific responses. Here we identify context geometry compression (CGC), in which models preserve cellular identity and shared perturbation effects while collapsing differences in how the same perturbation acts across contexts. Across biological systems, reproducible context-specific responses remained difficult to recover; in a large drug-response matrix, this persisted even with nearly complete perturbation coverage. The same drug-response predictions achieved higher normalized recovery in predefined pathway summaries than in gene-level response profiles, while coarse response programs were more predictable in human lymphoblastoid cell lines. Compact baseline transcriptomic representations retained most full-RNA predictive performance for patient-specific ex vivo drug responses. Together, these findings establish recoverable biological resolution, rather than measurement dimensionality, as a criterion for evaluating virtual cell predictions and matching model outputs to the biological detail supported by evidence.
Related
- RELAY: Relational In-Context Learning for Cellular Perturbation Prediction
- scDisentangle: sequential disentanglement for single-cell 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
- MultiFlow: coupled flow matching for predicting single-cell multiomic perturbation responses in unseen cellular contexts
BibTeX
@misc{huang2026recoverable,
title = {{The recoverable resolution of cellular perturbation-response prediction}},
author = {Y. Huang and H. Wang and C. Li and P. C. Wilson},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.30.755756},
url = {https://www.biorxiv.org/content/10.64898/2026.09.30.755756v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - The recoverable resolution of cellular perturbation-response prediction
AU - Y. Huang
AU - H. Wang
AU - C. Li
AU - P. C. Wilson
PY - 2026
DA - 2026/09/30
PB - bioRxiv
DO - 10.64898/2026.09.30.755756
UR - https://www.biorxiv.org/content/10.64898/2026.09.30.755756v1
AB - Virtual cell models aim to predict how cells respond to perturbations, yet accurate prediction of post-perturbation gene expression can conceal failures to recover context-specific responses. Here we identify context geometry compression (CGC), in which models preserve cellular identity and shared perturbation effects while collapsing differences in how the same perturbation acts across contexts. Across biological systems, reproducible context-specific responses remained difficult to recover; in a large drug-response matrix, this persisted even with nearly complete perturbation coverage. The same drug-response predictions achieved higher normalized recovery in predefined pathway summaries than in gene-level response profiles, while coarse response programs were more predictable in human lymphoblastoid cell lines. Compact baseline transcriptomic representations retained most full-RNA predictive performance for patient-specific ex vivo drug responses. Together, these findings establish recoverable biological resolution, rather than measurement dimensionality, as a criterion for evaluating virtual cell predictions and matching model outputs to the biological detail supported by evidence.
ER -