tspcalc                package:tspair                R Documentation

_E_s_t_i_m_a_t_e _t_o_p _s_c_o_r_i_n_g _p_a_i_r_s _f_r_o_m _a _g_e_n_e _e_x_p_r_e_s_s_i_o_n _m_a_t_r_i_x

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

     This function calculates the pair of genes that show the maximum
     difference in ranking between two user specified groups. The "top
     scoring pair" (TSP) maximizes the average of sensitivity and
     specificity over all rank based classifiers using a pair of genes
     in the data set.

_U_s_a_g_e:

       tspcalc(dat,grp)

_A_r_g_u_m_e_n_t_s:

     dat: Can take two values: (a) an m genes by n arrays matrix of
          expression data or (b) an eSet object

     grp: Can take one of two values: (a) A group indicator in
          character or numeric form, (b) an integer indicating the
          column of pData(dat) to use as the group indicator.

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

     tspcalc only works for two group classification. The computation
     time grows rapidly in the number of  genes, so for large gene
     expression matrices one should be prepared to wait or do a
     pre-filtering step. The top scoring pairs methodology was
     originally described in Geman et al. (2004) and the unique TSP
     idea was described in Tan et al. (2005).

_V_a_l_u_e:

     A tsp object with elements: 

   index: A two-column matrix where each row is a pair of indices from
          the data matrix achieving the top score

tspscore: The rank based score described in Geman et al. (2004),
          essentially the empirical average of sensitivity and
          specificity for the pair.

   score: The tie-breaking score described in Tan et al. (2005).

     grp: The binary group indicator.

  tspdat: Row i and row (i + Number of total TSPs) of this data matrix
          represent the expression data for the ith TSP pair.

  labels: The group labels for the two groups as defined by the grp
          variable.

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

     Jeffrey T. Leek jtleek@jhu.edu

_R_e_f_e_r_e_n_c_e_s:

     D. Geman, C. d'Avignon, D. Naiman and R. Winslow, "Classifying
     gene expression profiles from pairwise mRNA comparisons," 
     Statist. Appl. in Genetics and Molecular Biology, 3, 2004.

     A.C. Tan, D.Q. Naiman, L. Xu, R.L. Winslow, D. Geman, "Simple
     decision rules for classifying human cancers from gene expression
     profiles," Bioinformatics, 21: 3896-3904, 2005.

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

     'tspplot', 'ts.pair', 'tspsig','predict.tsp', 'summary.tsp'

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

       ## Not run: 
       ## Load data
       data(tspdata) 

       ## Run tspcalc() on a data matrix and grp vector
       tsp1 <- tspcalc(dat,grp)
       tsp1$index

       ## Run tspcalc() on an expression set and a column of the pData matrix
       tsp2 <- tspcalc(eSet1,1)
       tsp2$index
      ## End(Not run)

