netReg

The software tool 'netReg' is a package for graph-penalized regression models that utilize prior knowledge on dependencies in the form of graphs to model biological associations or dependencies using linear regression. The package is implemented as both an R-package and a C++ command-line tool and can handle high-dimensional and multivariate genomic datasets with large networks and thousands of variables. The main computations are done in C++, using Armadillo for fast matrix calculations and Dlib for optimization. The package yields sparse or smooth solutions for regression coefficients.