Continuous Topological Accessibility (RFA) as a 3D-Aware GRN Filter: Anticorrelation with eQTL and No Prediction of ΔRNA

RFA computed from Hi-C–based Markov accessibility anticorrelates with eQTL in GTEx v10 LCL (chr1+17+19, 10 kb), with AUC 0.386 and rho −0.09.

• Why it matters: This addresses whether a 3D-aware topological accessibility score predicts gene regulation, and whether any signal reflects local contact distance-decay versus higher-order chromatin structure.
• What they did: They compute RFA by solving (I − {gamma}P)x = v0 using TF or ATAC peaks, then test association with eQTL and {triangleup}RNA across specified genomic windows and gene sets.
• The result: The eQTL anticorrelation is reproduced by distance-preserving Hi-C shuffles (AUC shift within [−0.074, +0.076]) but destroyed by degree-preserving rewiring, while RFA shows no prediction of {triangleup}RNA (rho +0.005, p = 0.87).

AI summary of the abstract · may contain errors · not medical advice · read the paper.