Network reconstruction with local partial correlation: comparative evaluation
share
Over the past decade, various methods have been proposed for the reconstruction of networks modeled as Gaussian Graphical Models. In this work we analyzed three different approaches: the Graphical Lasso (GLasso), Graphical Ridge (GGMridge) and Local Partial Correlation (LPC). For the evaluation of the methods, we used high dimensional data generated from simulated random graphs (Erdös-Rényi, Barabási-Albert, Watts-Strogatz). The performance was assessed through the Receiver Operating Characteristic (ROC) curve. In addition the methods were used for reconstruction of co-expression network, for differentially expressed genes in human cervical cancer data. LPC method outperformed the GLasso in most of the simulation cases, even though GGMridge produced better ROC curves then both other methods. LPC obtained similar outcomes as GGMridge in real data studies.
READ FULL TEXT
