makeStepList            package:stepNorm            R Documentation

_C_o_n_s_t_r_u_c_t_i_o_n _o_f _a _s_t_e_p_w_i_s_e _n_o_r_m_a_l_i_z_a_t_i_o_n _l_i_s_t

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

     This function provides a user friendly way to construct a list for
     input to the function 'stepWithinNorm'. The list indicates
     intended biases for correction and models for stepwise
     normalization.

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

       makeStepList(A = c("median", "rlm", "loess"), PT = c("median", "rlm",
     "loess"), PL = c("median", "rlm", "loess"), Spatial2D = c("rlm2D",
     "loess2D", "aov2D", "spatialMedian"))

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

       A: A character string specifying the normalization models for
          the adjustment of intensity or A bias:

          _m_e_d_i_a_n:  global median location normalization

          _r_l_m:  global intensity or A-dependent robust linear
               normalization using the 'rlm' function

          _l_o_e_s_s:  global intensity or A-dependent robust nonlinear
               normalization using the 'loess' function

          The user can specify any of these three choices and the
          selected model will be compared based the goodness fit and
          model parsimony; If the correction of the $A$ bias is not
          desired, the user can set 'A = NULL'. 

      PT: A character string specifying the normalization models for
          the adjustment of print-tip or PT bias:

          _m_e_d_i_a_n:  within-print-tip-group median normalization

          _r_l_m:  within-print-tip-group robust linear normalization
               using the 'rlm' function

          _l_o_e_s_s:  within-print-tip-group robust nonlinear normalization
               using the 'loess' function

          _n_o_n_e:  no normalization for the PT bias

          If the correction of the $PT$ bias is not desired, the user
          can set 'PT = NULL'. 

      PL: A character string specifying the normalization models for
          the adjustment of well-plate or PL bias:

          _m_e_d_i_a_n:  within-well-plate median normalization

          _r_l_m:  within-well-plate robust linear normalization using the
               'rlm' function

          _l_o_e_s_s:  within-well-plate robust nonlinear normalization
               using the 'loess' function

          _n_o_n_e:  no normalization for the PL bias

          If the correction of the $PL$ bias is not desired, the user
          can set 'PL = NULL'. 

Spatial2D: A character string specifying the normalization models for
          the adjustment of spatial 2D bias:

          _n_o_n_e:  no normalization for the spatial 2D bias

          _a_o_v_2_D:  spatial bivariate location normalization using ANOVA

          _r_l_m_2_D:  spatial bivariate location normalization using the
               'rlm' function

          _l_o_e_s_s_2_D:  spatial bivariate location normalization using the
               'loess' function

          _s_p_a_t_i_a_l_M_e_d_i_a_n:  spatial location normalization using a
               spatial median approach (see Wilson et al. (2003) in
               reference)

          If the correction of the $PL$ bias is not desired, the user
          can set 'Spatial2D = NULL'. 

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

     This function provides a user friendly way to specify the
     parameter 'wf.loc' for the main stepwise normalization function
     'stepWithinNorm'; see examples for details.

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

     An object of class "list" for input to the 'stepWithinNorm'
     function.

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

     Yuanyuan Xiao, yxiao@itsa.ucsf.edu, 
      Jean Yee Hwa Yang, jean@biostat.ucsf.edu

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

     'stepWithinNorm'.

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

     # Examples use swirl dataset, for description type ? swirl
     data(swirl)

     # To use the default parameters, which adjusts A, PT, PL and Spatial2D
     # biases sequentially and compares all models available, simple type

     wf.loc <- makeStepList()

     # To apply loess for the A bias, and to omit the Spatial2D step
     wf.loc <- makeStepList(A=("loess"), Spatial2D=NULL)

     # To compare only rlm and loess in the A bias step, and other biases as default
     wf.loc <- makeStepList(A=c("rlm","loess"))

     # input to the stepWithinNorm function
     ## Not run: 
     step.swirl1 <- stepWithinNorm(swirl[,1],wf.loc=wf.loc)
     ## End(Not run)

