bhklab / RLOBICO

R implementation of Logic Optimization for Binary Input to Continuous Output (LOBICO)

RLOBICO

The logical models implemented in this package are based on the work of Knijnenburg et al. in:

Knijnenburg et al, Logic models to predict continuous outputs based on binary inputs with an application to personalized cancer therapy, Sci Rep 2016

System Requirements

1. This package is currently unsuported on Windows operating systems. It has been successfully tested on Ubuntu 18.04 and MacOS Mojave, but we expect other versions and distributions to be compatible.

2. C++11 is required to run included C/C++ functions. The package was developed and tested using GNU's g++ compiler.

3. Rcpp is required to compile the C functions for use in R. Please ensure you have the latest version if you encounter installation issues with this package.

4. IBM ILOG CPLEX is required to solve optimization problems in logical models built with this package. We have tested this package with IBM ILOG CPLEX 12.9; if you have install issues please update CPLEX. IBM ILOG CPLEX is free under IBM's academic initiative program for students and academic researchers. It can be downloaded here with a valid university associated email.

Installation

1. Currently this package is only available on Unix and MacOS due to compatability issues between IBM ILOG CPLEX and the Windows RTools compiler. We are currently trouble-shooting this issue and will update the README if any progress is made.

2. Please install the IBM ILOG CPLEX Optimization Studio (v>=12.9.0). The academic version is highly recommended, large size problems (>1,000 variables, >1,000 equations) will not be solved under the community version. The exception "Bad promotional version, problem size exceeds limit" will be raised.

3. The included configure script should be sufficient to locate your IBM ILOG CPLEX installation directory and create the src/Makevars file for your installation. If this process is unsuccessful we recommend moving the configure script out of the RLOBICO directory and manually creating a Makevars file. We have included examples.Makevars with the path and flags for the default installation locations on Ubuntu and MacOS. If these do not work you will need find the appropriate directories and set the package paths in Makevars accordingly.

4. This package can be installed from GitHub with devtools::install_github("bhklab/RLOBICO"). We are also in the process of submitting to CRAN, after which the package will be available using the standard install.packages function.

Example Data

The BIBW2992 dataset is included in this package to power our examples and vignettes.

Trouble Shooting

1. If you receive errors related to missing .h files, this is likely due to use of an incompatible C compiler. We recommend using a recent version of g++, specifically one that supports gnu++11 or c++11. Although this compiler should be the default on linux machines, if it is not you can download and install it using sudo apt install. On MacOS we recommend using Xcode's command line tools; installation instructions can be found here.

2. If the configure script is unable to locate your CPLEX installation, and you do not wish to manually configure your Makevars file from example.Makevars, you can uninstall CPLEX, redownload it as listed above, and install it to the DEFAULT directory suggested by the installer. This should resolve the issue. If it does not please check the version of your CPLEX install as well as other system dependencies listed in DESCRIPTION