CDS: Counterfactual Directionality Score for Structured Interventions in Spatial Graphs
For researchers analyzing spatial biological systems, this provides a principled method to infer directional cell-cell interactions from graph-based models, though it is an incremental improvement over existing correlation-based approaches.
The paper introduces the Counterfactual Directionality Score (CDS) to quantify directional influence between node populations in spatial graphs, addressing the lack of principled evaluation under controlled perturbations. Experiments on synthetic data show CDS recovers directional influence and is robust to confounding, with preliminary spatial transcriptomics results revealing biologically plausible interactions.
Quantifying directional influence between node populations is a fundamental problem in graph-based modeling, particularly in spatial biological systems where cell-cell interactions shape functional outcomes. Existing approaches based on attention, attribution, or correlation capture associations but do not provide a principled framework for evaluating directional effects under controlled perturbations. We introduce a framework for structured counterfactual interventions in graph-based models to estimate directional influence between node types. Our approach trains a Neighbor Influence Model (NIM) to predict node states from local neighborhoods and applies constrained interventions that modify neighborhood composition while preserving key spatial and structural properties. We define the Counterfactual Directionality Score (CDS), which measures the change in predicted node state induced by targeted perturbations, and provide a theoretical interpretation of CDS as a finite-difference measure of local intervention sensitivity. To obtain valid uncertainty estimates, we introduce a core-level bootstrap procedure that accounts for dependencies within spatial samples. Experiments on synthetic spatial graphs with known directional structure show that CDS recovers directional influence, remains well calibrated under null conditions, and is robust to confounding signals, while preliminary results on spatial transcriptomics data reveal biologically plausible and consistent interactions across tissue cores.