PIE: Generalizing perturbation effects across unseen perturbations, contexts and datasets
- 1Arc Institute
Abstract
Predicting cellular response to perturbations is key to understanding biological mechanisms and selecting therapeutic targets. However, generalizing across cellular contexts, perturbations, and experimental datasets remains challenging due to incomplete representations of the system being perturbed and the difficulty of measuring true perturbation effects within that system. We introduce PIE, a model that addresses these limitations by reformulating the learning task around population level perturbation effects and incorporating auxiliary inputs that better characterize the biological system. PIE directly predicts differentially expressed genes and changes in gene expression for each (context, perturbation) pair, using external biological knowledge, baseline gene expression, and observed perturbation responses. Its architecture accommodates input sources with variable feature sets and predicts effects on genes seen and unseen during training. On the Replogle Nadig dataset, PIE achieves state of the art performance on most metrics across all tested generalization settings, including the most challenging setting where both the context and perturbation are unseen during training. For predicting differentially expressed genes, PIE achieves 1.2 to 3.2 times the AUPRC of the strongest baseline in each setting: across unseen contexts, perturbations, experimental datasets, and jointly unseen contexts and perturbations. Overall, PIE provides a framework for generalizing perturbation effects across diverse experimental conditions.
Related
- ExPert infers causal perturbations from cell states across biological and clinical contexts
- Pop-Corn: Predicting Perturbation Phenotype Effects Across Single-Cell and Spatial Contexts
- MultiFlow: coupled flow matching for predicting single-cell multiomic perturbation responses in unseen cellular contexts
- Generalizing Protein Fitness Prediction Across Distribution Shifts via Structural and Co-evolutionary Priors
- Protocol for Discovering Convergent Molecular Networks across Multi-Omics Datasets Using Network Propagation
BibTeX
@misc{verma2026generalizing,
title = {{PIE: Generalizing perturbation effects across unseen perturbations, contexts and datasets}},
author = {R. Verma and A. Adduri and B. Bevilacqua and B. Eraslan and D. Burke and H. Goodarzi and Y. Roohani},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.02.756297},
url = {https://www.biorxiv.org/content/10.64898/2026.10.02.756297v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - PIE: Generalizing perturbation effects across unseen perturbations, contexts and datasets
AU - R. Verma
AU - A. Adduri
AU - B. Bevilacqua
AU - B. Eraslan
AU - D. Burke
AU - H. Goodarzi
AU - Y. Roohani
PY - 2026
DA - 2026/10/02
PB - bioRxiv
DO - 10.64898/2026.10.02.756297
UR - https://www.biorxiv.org/content/10.64898/2026.10.02.756297v1
AB - Predicting cellular response to perturbations is key to understanding biological mechanisms and selecting therapeutic targets. However, generalizing across cellular contexts, perturbations, and experimental datasets remains challenging due to incomplete representations of the system being perturbed and the difficulty of measuring true perturbation effects within that system. We introduce PIE, a model that addresses these limitations by reformulating the learning task around population level perturbation effects and incorporating auxiliary inputs that better characterize the biological system. PIE directly predicts differentially expressed genes and changes in gene expression for each (context, perturbation) pair, using external biological knowledge, baseline gene expression, and observed perturbation responses. Its architecture accommodates input sources with variable feature sets and predicts effects on genes seen and unseen during training. On the Replogle Nadig dataset, PIE achieves state of the art performance on most metrics across all tested generalization settings, including the most challenging setting where both the context and perturbation are unseen during training. For predicting differentially expressed genes, PIE achieves 1.2 to 3.2 times the AUPRC of the strongest baseline in each setting: across unseen contexts, perturbations, experimental datasets, and jointly unseen contexts and perturbations. Overall, PIE provides a framework for generalizing perturbation effects across diverse experimental conditions.
ER -