Partial Correlation Coefficient with Information Theory

Apply Partial Correlation coefficient with Information Theory (PCIT) to a correlation matrix. The PCIT algorithm identifies meaningful correlations to define edges in a weighted network. The algorithm can be applied to any correlation-based network including but not limited to gene co-expression networks. To reduce compute time by making use of multiple compute cores, simply run PCIT on a computer with has multiple cores and also has the Rmpi package installed. PCIT will then auto-detect the multicore environment and run in parallel mode without the need to rewrite your scripts. This makes scripts, using PCIT, portable across single core (or no Rmpi package installed) computers which will run in serial mode and multicore (with Rmpi package installed) computers which will run in parallel mode.

Build Status

PCIT: an R package for weighted gene co-expression networks based on partial correlation and information theory approaches

Building and Submitting Packages

  1. Update "Version" and "Date" in the DESCRIPTION file
  2. Add info about changes since last release to ChangeLog file
  3. R CMD build pcit
  4. Check the built tarball: R CMD check --as-cran PCIT_*.tar.gz
  5. Submit the tarball at or via FTP uisng instructions here:
  6. Tag the release on github:
version=$(grep "^Version:" DESCRIPTION | awk '{print $2}')
# Tag the current branch using the Changelog entry for this version, asthe message
git tag -a "v${version}" -F <(perl -e '$v_encountered=0; while(<>){if (/'${version}'/){$v_encountered=1;print;next}; last if /^\d/; print}' ChangeLog)
# Push the tag up to github
git push origin "v${version}"


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.5-3 by Nathan S. Watson-Haigh, 5 years ago

Browse source code at

Authors: Nathan S. Watson-Haigh

Documentation:   PDF Manual  

GPL-3 license

Suggests Rmpi

Suggested by networkABC.

See at CRAN