Benchmarking precision matrix estimation methods for differential co-expression network analysis

Abstract Background Gene expression profiling is widely used to investigate disease mechanisms, but classical approaches such as differential expression or pairwise correlation analyses provide limited interpretability. Network-based differential co-expression methods that model conditional dependencies through partial correlations offer richer insights, yet their application in high-dimensional settings requires estimation of precision matrices. Numerous precision matrix estimation methods (PMEMs) have been proposed, but these are typically validated in a narrow range of simulation settings and benchmarked against only a small number of competing methods, so their relative performance under diverse conditions remains unclear. Results Simulated gene expression datasets with known ground truth correlation structures were used to benchmark a broad set of PMEMs for their suitability in differential co-expression network analysis of transcriptomic data. Performance was strongly affected by data characteristics, including covariance structure, matrix density, covariance values, sample size-to-dimension ratio, and sampling distribution. Among the evaluated methods, GLassoElnetFast generally achieved high accuracy in recovering differential edges across the simulated scenarios, although high signal-to-noise ratios and sufficient sample sizes remain essential for reliable inference. Applying a subset of these methods to a real transcriptomic dataset reproduced this estimator-dependent variability, with limited overlap between the differential edges recovered by different methods. Conclusions Evaluation across diverse simulation conditions demonstrated that no single metric or condition was sufficient to assess PMEM performance. Therefore, less extensive evaluations may only capture part of this variability and limit the generalizability of their conclusions. Our simulation and benchmarking framework supports future method development and ensures reproducible evaluation of newly developed approaches.