\import{mcx.zmm} \begin{pud::man}{ {name}{clm mate} {html_title}{The clm mate manual} {author}{Stijn van Dongen} {section}{1} {synstyle}{long} {defstyle}{long} \man_share } \${html}{\"pud::man::maketoc"} \sec{name}{NAME} \NAME{clm_mate}{compute best matches between two clusterings} \disclaim_clm{mate} \sec{synopsis}{SYNOPSIS} \par{ \clm{mate} \synoptopt{-o}{fname}{output file name} \synoptopt{-b}{omit headers} \synoptopt{--one-to-many}{require multiple hits in } \stdsynopt } \sec{description}{DESCRIPTION} \par{ \clm{mate} computes for each cluster \v{X} in \v{clfile1} all clusters \v{Y} in \v{clfile2} that have non-empty intersection and outputs a line with the data points listed below.} \verbatim{\:/ overlap(X,Y) # 2 * size(meet(X,Y)) / (size(X)+size(Y)) index(X) # name of cluster index(Y) # name of cluster size(meet(X,Y)) size(X-Y) # size of left difference size(Y-X) # size of right difference size(X) size(Y) projection(X, clfile2) # see below projection(Y, clfile1) # see below } \par{ The projected size of a cluster \v{X} relative to a clustering \v{K} is simply the sum of all the nodes shared between any cluster \v{Y} in \v{K} and \v{X}, duplications allowed. For example, the projected size of \v{(0,1)} relative to \v{\{(0,2,4), (1,4,9), (1,3,5)\}} equals \v{3}.} \par{ The overlap between \v{X} and \v{Y} is exactly 1.0 if the two clusters are identical, and for nearly identical clusterings the score will be close to 1.0.} \par{ All of this information can also be obtained from the contingency matrix defined for two clusterings. The \v{[i,j]} row-column entry in a contigency matrix between to clusterings gives the number of entries in the intersection between cluster\~\v{i} and cluster\~\v{j} from the respective clusterings. The other information is implicitly present; the total number of nodes in clusters\~\v{i} and\~\v{j} for example can be obtained as the sum of entries in row\~\v{i} and column\~\v{j} respectively, and the difference counts can then be obtained by subtracting the intersection count. The contingency matrix can easily be computed using \mcx; e.g.} \verbatim{ mcx /clfile2 lm /clfile1 lm tp mul /ting wm} \car{ will create the contingency matrix in mcl matrix format in the file \v{ting}, where columns range over the clusters in \v{clfile1}.} \par{ The output can be put to good use by sorting it numerically on that first score field. It is advisable to use a stable sort routine (use the \usearg{-s} option for UNIX sort) From this information one can quickly extract the closest clusters between two clusterings.} \sec{options}{OPTIONS} \'begin{itemize}{\mcx_itemopts} \item{\defopt{-o}{fname}{output file name}} \car{Specify the name of the output file.} \item{\defopt{-b}{omit headers}} \car{Batch mode, omit column names.} \item{\defopt{--one-to-many}{require multiple hits in }} \car{Do not output information for clusters in the first file that are subset of a cluster in the second file.} \stddefopt \'end{itemize} \sec{author}{AUTHOR} \par{ Stijn van Dongen. } \sec{seealso}{SEE ALSO} \par{ \mysib{mclfamily} for an overview of all the documentation and the utilities in the mcl family. } \end{pud::man}