evaporative-cooling

IntroductionEvaporative cooling (EC) feature selection is a command-line data mining software implementation in Java and Fortran for filtering genetic association data. EC integrates Random Forests and Relief-F attribute importance measures in order to balance independent and interaction effects while removing attributes that are irrelevant to the phenotype. EC has been tested on single-nucleotide polymorphism (SNP) data. For those with access to a cluster, the parallel version of EC will be released soon.

New Windows GUI version

Quick Start for *nix command line

FAQs

RequirementsJava, Fortran 90 compiler (f95, ifort, or the like). EC has been tested on Linux systems. The code should work on Windows or Mac given a fortran compiler.

ReferencesB.A. McKinney, D. M. Reif, B. C. White, J. E. Crowe Jr., J. H. Moore. "Evaporative cooling feature selection for genotypic data involving interactions." Bioinformatics 2007, 23: 2113-2120. pdf

B.A. McKinney, J.E. Crowe, Jr., J. Guo, and D. Tian, "Capturing the spectrum of interaction effects in genetic association studies by simulated evaporative cooling network analysis," 2009, PLoS Genetics 2009, 5(3): e1000432. doi:10.1371/journal.pgen.1000432. open source

AcknowledgmentsWe would like to acknowledge the developers of PARF (parallel Random Forests, written in Fortran 90) and WEKA (data mining software written in Java).

Send inquiries to brett.mckinney@gmail.com.

McKinneyLab

Tags

  fortran genetics java data-mining bioinformatics machine-learning

Share