classificationModel       package:gene2pathway       R Documentation

_H_i_e_r_a_r_c_h_i_c_a_l _C_l_a_s_s_i_f_i_c_a_t_i_o_n _M_o_d_e_l

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

     This file contains the hierarchical classification model to
     predict KEGG pathway branches for genes. The model uses a pruned
     KEGG hierarchy, where metabolic pathways are not distinguished
     further, and the KEGG hierarchy for "cellular processes" and
     "genetic information processing" is pruned at the 2nd level. By
     default the model uses bagging to improve prediction accuracy.
     Important: There exists one separate model file for each organism.

_F_o_r_m_a_t:

     List of class "model", where each model has the following entries:

     _W learned decision hyperplane normal vector

     _C dictionary of label vectors, which can be predicted individually
          or which can be used to predict combinations of them

     _d_e_t_e_c_t_o_r_s SVM models trained to separate one specific pathway
          branch from the rest of the hierarchy

     _u_s_e_d__d_o_m_a_i_n_s InterPro domains used by the classifier to separate
          the specific branch from the rest of the hierarchy

     _a_l_l_d_o_m_a_i_n_s all InterPro domains used to build feature vectors

     _a_l_l_p_a_t_h_w_a_y_s hierarchy branches, which can be predicted

     _t_r_e_e_s_i_z_e_s relative size of hierarchy below the corresponding
          branch

     _k_e_g_g__h_i_e_r_a_r_c_h_y a nested list with information (parent branches,
          pathway names, pathway IDs, hierarchy level) on all higher
          hierarchy branches for each pathway

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

     Holger Froehlich

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

     'classificationModelSignalTrans'

