Skip to content

DataFusion4NetBio/RF4Bioinformatics

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 

Repository files navigation

RF4Bioinformatics

Random Forest Wrapper Code for book chapter "random forest for Bioinformatics"


If you use this code, please cite :

  • Major paper:

@incollection{qi2012random, title={Random forest for bioinformatics}, author={Qi, Yanjun}, booktitle={Ensemble Machine Learning}, pages={307--323}, year={2012}, publisher={Springer} }


PDF @ http://www.cs.cmu.edu/%7Eqyj/papersA08/11-rfbook.pdf

  • Related Papers,

Y. Qi, HK. Dhiman, et al, Z. Bar-Joseph, J. Klein-Seetharaman,(2009) "Systematic prediction of human membrane receptor interactions" PROTEOMICS 2009, 9, 5243-5255



I assume that you have the "g77" command in your command list. (most unix machines have this installed.)

If not, for windows, you can install the software: "MinGW". Remember to add the g77 in your windows command list after installations.


It is very easy to run this code:

  1. Just put your parameter in a parameter file for example: testpara.file totally 10 parameters (all related files' names should be also in the input parameter file.)

  2. Then run the perl wrapper perl change_RF_codeRunParaF.pl testpara.file


I assume that your input feature files have been pre-processed ==> which means they contain all real features and the features have no missing values.


In the subdirectory, there exsit the example files configurying in the "testpara.file"


perl wrapper "process_RFtestOut.pl" is an extra script. Since the RF output contains the voting from all trees about postive leaves or negative leaves.. Thus one summary score could be just the ==> positive vote score - negative vote score

This wrapper would convert the direct output RF file into the summary score file as described.

Releases

No releases published

Packages

No packages published