Methodology for supervised grouping aka "clustering" of potentially many predictor variables, such as genes etc, implementing algorithms 'PELORA' and 'WILMA'.
supclust is an R package which has been on
CRAN since January 2004, see (https://cran.r-project.org/package=supclust).
It contains C and R code, written by Marcel Dettling in 2003, implementing the WILMA and PELORA algorithms for supervised clustering of potentially many predictor variables, notably for gene modeling.
Martin Maechler who had helped originally in getting the R package to CRAN had taken maintenance in 2009, to prevent long term orphaning of the package.
The source code has been moved to github only on Oct 15, 2020, using my own
G2RCSn utility and notably the
rcs-fast-export.rb ruby script,
see also my
blog post
about how I achieved such a transformation for my
sfsmisc CRAN R package.