clusterAlgorithmClara-class    package:goCluster    R Documentation

_C_l_a_s_s "_c_l_u_s_t_e_r_A_l_g_o_r_i_t_h_m_C_l_a_r_a", _p_r_o_v_i_d_e_s _p_a_r_t_i_t_i_o_n_i_n_g _a_r_o_u_n_d
_m_e_d_o_i_d_s _o_f _l_a_r_g_e _d_a_t_a_s_e_t_s

_D_e_s_c_r_i_p_t_i_o_n:

     This can be used to group a dataset according to partitioning
     around medoids. The resulting gene groups can  subsequently be
     analysed for significant enrichment of specific annotations.

_D_e_t_a_i_l_s:

     The class provides a wrapper around the 'clusterclara' function.
     Please read the corresponding documentation for further details.

_S_l_o_t_s:

     '_c_l_u_s_t_e_r_s': Object of class '"numeric"', determines the number of
          clusters the partitioning around medoids will identify.

     '_r_e_p_e_a_t_s': Object of class '"numeric"', specifies how often the
          clustering is repeated in case clara is not run with a fixed
          initialization.

     '_f_i_x_e_d': Object of class '"logical"', if true, a fixed seed will
          be used for the partitioning around medoids. 

     '_d_i_s_t_a_n_c_e': Object of class '"character"', specifies the distance
          matrix that will be used.

     Additional slots are described in the documentation of the
     'clusterAlgorithm-class' and 'clusterModule-class'.

_E_x_t_e_n_d_s:

     Class '"clusterAlgorithm"', directly. Class '"clusterModule"', by
     class '"clusterAlgorithm"'.

_M_e_t_h_o_d_s:

     _c_o_n_f_i_g 'signature(object = "clusterModule")': interactive setup of
          the class. You will be asked to specify the number of
          clusters clara clustering should result in and whether a
          fixed seed should be used. If not the class offers to repeat
          the clustering. In addition the distance matrix needs to be
          defined.

     _s_e_t_u_p 'signature(object = "clusterModule")': returns the
          configuration of the object as a list. This list can again be
          used for the non-interactive setup of the class. 

     _s_e_t_u_p<- 'signature(object = "clusterModule")': non-interactive
          setup of the class. The options are specified using a list. 

     _e_x_e_c_u_t_e 'signature(object = "clusterModule")': run the clustering. 

     _r_e_s_e_t 'signature(object = "clusterModule")': remove all cluster
          data so that the 'execute' function can be run again.

_A_u_t_h_o_r(_s):

     Gunnar Wrobel, <URL: http://www.gunnarwrobel.de>.

_S_e_e _A_l_s_o:

     'clusterclara', 'goCluster-class', 'clusterModule-class',
     'clusterAlgorithm-class', 'clusterAlgorithmKmeans-class',
     'clusterAlgorithmPam-class', 'clusterAlgorithmHclust-class', .

_E_x_a_m_p_l_e_s:

     ## Predefined setup for goCluster
     ## (This configuration selects the
     ## clara clustering)
     data(benomylsetup)

     ## Setup a new goCluster object
     test <- new("goCluster")
     setup(test) <- benomylsetup

     ## Retrieve annotation
     test@data <- execute(test@data, test)

     ## Cluster the dataset
     test@algo <- execute(test@algo, test)

