[2603.03493] Quantifying Ranking Instability Across Evaluation Protocol Axes in Gene Regulatory Network Benchmarking

Quantitative Biology > Molecular Networks arXiv:2603.03493 (q-bio) [Submitted on 3 Mar 2026] Title:Quantifying Ranking Instability Across Evaluation Protocol Axes in Gene Regulatory Network Benchmarking Authors:Ihor Kendiukhov View a PDF of the paper titled Quantifying Ranking Instability Across Evaluation Protocol Axes in Gene Regulatory Network Benchmarking, by Ihor Kendiukhov View PDF HTML (experimental) Abstract:Benchmark rankings are routinely used to justify scientific claims about method quality in gene regulatory network (GRN) inference, yet the stability of these rankings under plausible evaluation protocol choices is rarely examined. We present a systematic diagnostic framework for measuring ranking instability under protocol shift, including decomposition tools that separate base rate effects from discrimination effects. Using existing single cell GRN benchmark outputs across three human tissues and six inference methods, we quantify pairwise reversal rates across four protocol axes: candidate set restriction (16.3 percent, 95 percent CI 11.0 to 23.4 percent), tissue context (19.3 percent), reference network choice (32.1 percent), and symbol mapping policy (0.0 percent). A permutation null confirms that observed reversal rates are far below random order expectations (0.163 versus null mean 0.500), indicating partially stable but non invariant ranking structure. Our decomposition reveals that reversals are driven by changes in the relative discrimination ability ...