This manual page is provisionary. It is a simple copy / paste of the manual of the command line tool.
NAME convert-classes DESCRIPTION Interconversions between different formats of class/cluster files. AUTHORS Jacques van Helden
Sylvain Brohée CATEGORY util USAGE convert-classes [-i inputfile] [-o outputfile] [-v] INPUT/OUTPUT FORMATS mcl The output format of the MCL algorithm. MCL is a graph-based clustering algorithm developed by Stijn Van Dongen (http://micans.org/mcl/). The output format has one row per cluster, each row giving a list of elements, separated by tabulations. mcode The output format of the MCODE algorithm. MCODE is a graph-based clustering algorithm developed by Bader and Hogue (Bader and Hogue, BMC Bioinformatics, 2003) http://baderlab.org/Software/MCODE tab A tab-delimited text file with 2 columns (+1 optional). The first column indicates the element, the second column the class (cluster). The third column is optional, it can be used to indicate the score for the assignation of the element to the class (e.g. a posterior probability). Any real value is accepted as score. profiles A tab-delimited text file with a table containing one row per element, and one column per class. The first row starting with a '#' (the header row) indicates the class names. The first column contains the element names. The cells of the table contain either boolean or real values. Boolean values indicate the membership (1 if the element is member of the class, 0 otherwise). Real values can specify a score for the membership of the element to the class (e.g. posterior probability). ermg Erdros-Renyi Mixture of Graphs. format defined by Stephane Robin for his ERMG algorithm (submitted). rnsc RNSC format. format defined by Andrew King for his RNSC clustering algorithm. Each line of the file represents a clusters contains a list of integer representing the member. The end of the classification is indicated by '-1'. To convert from RNSC to another format, two files must thus be given : a cluster file (-i option) and a name file (-names option) which is a two colum tab delimited file containing the name of each corresponding id (See King et al, 2004). OPTIONS -v # Level of verbosity (detail in the warning messages during execution) -h Display full help message -help Same as -h -i inputfile If no input file is specified, the standard input is used. This allows to use the command within a pipe. -o outputfile If no output file is specified, the standard output is used. This allows to use the command within a pipe. -names file_name Two column file specifying the labels of the members of the classification given in the classification file. First column contains the identifier and second column the corresponding label. -from input_format Input format. Supported: tab, mcl, profiles, mcode, rnsc -to output_format Output format. Supported: tab, mcl,profiles -mcol Member column. Column containing the member names in the tab format (default 1). -ccol Class column. Column containing the class names in the tab format (default 1). -scol Score column. Column containing the scores in tab format. If not specified, scores are not defined. -null Null string used as score in the profile output for the undefined class memberships (default 0). -all_scores Assign each node to all the clusters with the ERMG format, with the posterior probability as score. -min_score Minimal score value for member to class assignation. SEE ALSO