CFM-GP: Unified Conditional Flow Matching to Learn Gene Perturbation Across Cell Types

2025-08-08

AI Summary

This study addresses the critical limitation in genetic perturbation effect prediction—poor generalization across cell types and the necessity of cell-type-specific modeling. To overcome this, we propose the first cell-type-agnostic unified framework, which employs Conditional Flow Matching to model the continuous dynamic transformation from unperturbed to perturbed gene expression distributions. By elevating discrete perturbations to a differentiable, time-dependent continuous process, our approach enables scalable and biologically consistent inference. Cell-type-specific information is incorporated via conditional encoding, allowing parameter sharing across cell types while preserving biological fidelity. Evaluated on five single-cell datasets, our method achieves superior performance over state-of-the-art baselines in both R² and Spearman correlation. Pathway enrichment analysis further validates its high-fidelity recovery of biologically relevant signaling and regulatory pathways.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Game Theory and Economic Paradigms: Imperfect Information

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Understanding gene perturbation effects across diverse cellular contexts is a central challenge in functional genomics, with important implications for therapeutic discovery and precision medicine. Single-cell technologies enable high-resolution measurement of transcriptional responses, but collecting such data is costly and time-consuming, especially when repeated for each cell type. Existing computational methods often require separate models per cell type, limiting scalability and generalization. We present CFM-GP, a method for cell type-agnostic gene perturbation prediction. CFM-GP learns a continuous, time-dependent transformation between unperturbed and perturbed gene expression distributions, conditioned on cell type, allowing a single model to predict across all cell types. Unlike prior approaches that use discrete modeling, CFM-GP employs a flow matching objective to capture perturbation dynamics in a scalable manner. We evaluate on five datasets: SARS-CoV-2 infection, IFN-beta stimulated PBMCs, glioblastoma treated with Panobinostat, lupus under IFN-beta stimulation, and Statefate progenitor fate mapping. CFM-GP consistently outperforms state-of-the-art baselines in R-squared and Spearman correlation, and pathway enrichment analysis confirms recovery of key biological pathways. These results demonstrate the robustness and biological fidelity of CFM-GP as a scalable solution for cross-cell type gene perturbation prediction.

Problem

• Predict gene perturbation effects across diverse cell types
• Overcome limitations of separate models per cell type
• Learn continuous transformation between perturbed and unperturbed states

Innovation

• Unified model for gene perturbation prediction
• Flow matching captures perturbation dynamics
• Single model predicts across all cell types