RBioFS - A machine learning R package for random forest-based gene selection. This bioinformatic tool can confidently select miRNAs relevant to differentiating physiological phenotypes of interest. 

When using this program please cite: Zhang J, Hadj-Moussa H, and Storey KB. (2016). Current progress of high-throughput microRNA differential expression analysis and random forest gene selection for model and non-model systems: an R implementation. Journal of Integrative Bioinformatics, 13(5), 306.  PMID:28187420.

Download and install R and R Studio. To install the latest version of RBioFS, please select the following commands (copy/paste) and run them in R Studio (make sure you are connected to the internet):

(1) Run the following command to install devtools (needed to run RBioFS), otherwise skip this step:
(2) Run the following command to install Bioconductor (needed to run RBioFS), otherwise skip this step:
         For help with installing Bioconductor, visit (https://www.bioconductor.org/install/)
(3) Run the following command to install RBioFS
          devtools::install_github("jzhangc/git_RBioFS/RBioFS", repos = BiocInstaller::biocinstallRepos())
(4) To set a working directory use, replace "..." with your folder address:
         For Windows: setwd("working directory")    
         For Mac and Linux: setwd("working directory")
The source code and the R package are available on GitHub: https://github.com/jzhangc/git_RBioFS.git.

For a short RBioFS installation guide and user manual, download the following RBioFS User Manual .pdf

Sample input and output files  
The files below are examples of input and output files from the rbioFS machine learning gene selection function: 
DEMO- INPUT data format:  RBioFS Sample Data .csv  
DEMO- OUTPUT 1/5 - Imputed values: RBioFS Sample Data imputed .csv
DEMO- OUTPUT 2/5 - Initial feature selection: RBioFS Sample Data.initialFS .txt
DEMO- OUTPUT 3/5 - SFS-like selection: RBioFS Sample Data.SFS .txt
DEMO- OUTPUT 4/5 - Histogram ranking of miRNA VI values: RBioFS Sample Data.vi .plot .pdf
DEMO- OUTPUT 5/5 - OOB error rate (± SEM) change based on SFS-like selection: RBioFS_Sample_Data.OOB.plot .pdf