PMEM Evaluator

Precision Matrix Estimation Methods - Evaluator

This repository contains the simulation and evaluation framework used to benchmark precision matrix estimation methods (PMEMs) for differential co-expression analysis. The workflow generates condition-specific covariance structures, simulates data, across varying sample sizes, network densities, and covariance patterns. Performance is assessed using binary classification metrics (F1 Score, Accuracy, Matthews Correlation Coefficient (MCC)), matrix norms, Kullback-Leibler Losses, and differential edge recovery.

List of included PMEMs: glasso¹, clime², tiger³, quic⁴, bigquic⁵, equal⁶, scio⁷, GLassoElnetFast⁸, gdtrace⁹, rags2ridges¹⁰, fastclime, rope¹¹, glassofast, squic¹².

Running the Simulation/ Evaluation Pipeline

Option 1: prebuilt Docker image:

recommended/most reproducible option

docker pull matthias587/pmem-evaluator:2.2
docker run -it -v /path/to/config/files:/Configs -v /path/to/output/folder:/Output matthias587/pmem-evaluator:2.2 /Configs/configfile.yaml /Output

Option 2: compiling docker image

within the repo run

docker build -t your/imagename:1.0 .

Option 3: Source Code in R

within the repo, restore the renv using

R
renv::restore()

Rscript Pipeline/SimulationStudy.R Pipeline/Example_Configs/config1.yaml /path/to/output/folder

Important: for SQUIC to install/work you need to follow the set up described on https://www.gitlab.ci.inf.usi.ch/SQUIC

More information about the needed config file can be found in the ReadMe.md in the Pipeline Folder.

Footnotes

1. Friedman, J., Hastie, T., & Tibshirani, R. (2008). Sparse inverse covariance estimation with the graphical lasso. Biostatistics (Oxford, England), 9(3), 432–441. doi:10.1093/biostatistics/kxm045 ↩

2. Cai, T., Liu, W., & Luo, X. (2011). A Constrainedℓ1Minimization approach to sparse precision matrix estimation. Journal of the American Statistical Association, 106(494), 594–607. doi:10.1198/jasa.2011.tm10155 ↩

3. Liu, H., & Wang, L. (2012). TIGER: A tuning-insensitive approach for optimally estimating Gaussian graphical models. doi:10.48550/ARXIV.1209.2437 ↩

4. Hsieh, C.-J., Sustik, M. A., Dhillon, I. S., & Ravikumar, P. (2013). Sparse inverse covariance matrix estimation using quadratic approximation. doi:10.48550/ARXIV.1306.3212 ↩

5. URL: https://proceedings.neurips.cc/paper_files/paper/2013/file/1abb1e1ea5f481b589da52303b091cbb-Paper.pdf ↩

6. Wang, C., & Jiang, B. (2020). An efficient ADMM algorithm for high dimensional precision matrix estimation via penalized quadratic loss. Computational Statistics & Data Analysis, 142(106812), 106812. doi:10.1016/j.csda.2019.106812 ↩

7. Liu, W., & Luo, X. (2012). Fast and adaptive sparse precision matrix estimation in high dimensions. doi:10.48550/ARXIV.1203.3896 ↩

8. Kovács, S., Ruckstuhl, T., Obrist, H., & Bühlmann, P. (2021). Graphical Elastic Net and target matrices: Fast algorithms and software for sparse precision matrix estimation. doi:10.48550/ARXIV.2101.02148 ↩

9. Avagyan, V. (2022). Precision matrix estimation using penalized Generalized Sylvester matrix equation. Test (Madrid, Spain), 31(4), 950–967. doi:10.1007/s11749-022-00807-0 ↩

10. Peeters, C. F. W., Bilgrau, A. E., & van Wieringen, W. N. (2022). Rags2ridges: A one-stop- ℓ2 -shop for graphical modeling of high-dimensional precision matrices. Journal of Statistical Software, 102(4). doi:10.18637/jss.v102.i04 ↩

11. Kuismin, M. O., Kemppainen, J. T., & Sillanpää, M. J. (2017). Precision matrix estimation with ROPE. Journal of Computational and Graphical Statistics: A Joint Publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America, 26(3), 682–694. doi:10.1080/10618600.2016.1278002 ↩

12. Eftekhari, A., Gaedke-Merzhaeuser, L., Pasadakis, D., Bollhoefer, M., Scheidegger, S., & Schenk, O. (2021). Large-scale precision matrix estimation with SQUIC. SSRN Electronic Journal. doi:10.2139/ssrn.3904001 ↩