CombinedRanking-class      package:GeneSelector      R Documentation

"_C_o_m_b_i_n_e_d_R_a_n_k_i_n_g"

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

     Object returned from a call to 'GeneSelector'.

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


     '_r_a_n_k_i_n_g': 'Final' ranking, usually obtained by combinining
          several statistics. The generation of this ranking is
          described in GeneSelector.
           The first entry contains the index of the gene ranked
          highest, as in GeneRanking.

     '_r_a_n_k_m_a_t_r_i_x': Matrix of rankings, arranged in a way that the
          rankings from the most important statistic is in column 1. In
          contrast to 'ranking', the first row contains the rankings of
          the gene with the first _index_. This slot is used rather for
          internal reasons.

     '_i_n_o_u_t': Matrix arranged in the same way as 'rankmatrix', but
          information is now binary: If the specified threshold, then
          there is a '"+"' symbolizing selection, whereas a '"-"'
          symbolizes removal.

     '_s_e_l_e_c_t_e_d': The indices of those genes that fall below the
          specified threshold. Can be accessed more conveniently using
          'SelectedGenes'

     '_a_d_j_p_v_a_l': Numeric vector of adjusted p-values, ordered according
          to 'ranking'. 'NA' if no adjustment has been taken place (in
          the case that the threshold was fixed by the user).

     '_m_a_x_r_a_n_k': Threshold rank, either defined by the user or obtained
          via p-value adjustment.

     '_s_t_a_t_i_s_t_i_c_s': The names of the statistics used, ordered according
          to their importance  (as defined by the user).

     '_a_b_s_d_i_s_t': Absolute ('L1') distance from (theoretically) best
          possible result (rank 1 in all rankings), ordered according
          to 'ranking'. Note that minimum 'absdist' does not imply best
          rank and vice versa, because the computed distance does not
          weigh different statistics differently.

     '_r_e_l_d_i_s_t': A 'normalized' version of 'absdist' (lies in [0;1]).

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


     _s_h_o_w Use 'show(object)' for brief information.

     _t_o_p_l_i_s_t Use 'toplist(object, k=10)'   to get information about the
          top 'k=10' genes.
           The ranking used is a synthesis from several statistics.     

     _S_e_l_e_c_t_e_d_G_e_n_e_s Use 'SelectedGenes(object)' to show all genes that
          have been selected by the GeneSelector.

     _G_e_n_e_I_n_f_o_S_c_r_e_e_n Use 'GeneInfoScreen, which=1' to get detailed
          information about the gene with index 1, arranged in a pretty
          plot.

     _p_l_o_t Use 'plot(object)' to visualize relative distances, s.
          plot,CombinedRanking

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

     Martin Slawski martin.slawski@campus.lmu.de 
      Anne-Laure Boulesteix <URL: http://www.slcmsr.net/boulesteix>

