Fig. 7: Statistical evaluation of scaling characteristics.
From: A large-scale benchmark for network inference from single-cell perturbation data
Performance comparison in terms of Mean Wasserstein Distance (unitless; y-axis) of 10 methods for causal graph inference on observational data (top row; (see legend top right) and 11 methods on interventional scRNAseq data (top row; see legend top right) when varying the fraction of the full dataset size available for inference (in %; x-axis), and 11 methods on interventional data (bottom row; see legend bottom right) when varying the fraction of the full intervention set used (in %, x-axis). Markers indicate the values observed when running the respective algorithms with one of three random seeds, and colored lines indicate the median value observed across all tested random seeds for a method.
