Towards generalizable single-cell perturbation modeling via the Conditional Monge Gap
This addresses the challenge of generalizable perturbation modeling in single-cell biology, which is crucial for targeted therapies, though it appears incremental as an extension of neural optimal transport methods.
The paper tackles the problem of predicting single-cell responses to unseen treatments by proposing the Conditional Monge Gap, a method that learns optimal transport maps conditionally on covariates like drugs or dosages. It achieves results comparable or superior to condition-specific state-of-the-art methods on scRNA-seq and protein imaging data, with notable generalization to unseen drugs or dosages that outperforms other conditional models in capturing population heterogeneity.
Learning the response of single-cells to various treatments offers great potential to enable targeted therapies. In this context, neural optimal transport (OT) has emerged as a principled methodological framework because it inherently accommodates the challenges of unpaired data induced by cell destruction during data acquisition. However, most existing OT approaches are incapable of conditioning on different treatment contexts (e.g., time, drug treatment, drug dosage, or cell type) and we still lack methods that unanimously show promising generalization performance to unseen treatments. Here, we propose the Conditional Monge Gap which learns OT maps conditionally on arbitrary covariates. We demonstrate its value in predicting single-cell perturbation responses conditional to one or multiple drugs, a drug dosage, or combinations thereof. We find that our conditional models achieve results comparable and sometimes even superior to the condition-specific state-of-the-art on scRNA-seq as well as multiplexed protein imaging data. Notably, by aggregating data across conditions we perform cross-task learning which unlocks remarkable generalization abilities to unseen drugs or drug dosages, widely outperforming other conditional models in capturing heterogeneity (i.e., higher moments) in the perturbed population. Finally, by scaling to hundreds of conditions and testing on unseen drugs, we narrow the gap between structure-based and effect-based drug representations, suggesting a promising path to the successful prediction of perturbation effects for unseen treatments.