epistasis               package:happy               R Documentation

_A_n_a_l_y_s_i_s _o_f _E_p_i_s_t_a_s_i_s _b_e_t_w_e_e_n _M_a_r_k_e_r_s

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

     epistasis() will test for a statistical interaction between two
     sets of markers within the happy framework. The markers should be
     sufficiently far apart that they are unlinked (in practice 10cM
     for a 30 generation HS is sufficient). A partial F-test is
     performed to test if a model allowing for interactions fits better
     than a model in which each marker's contribution is additive
     between loci. Note that the effect of each marker within a locus
     can be either additive or full. Merging of strain is permitted.

     epistasispair() is the same as epistasis() except that only one
     pair of markers is tested.

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

     epistasis( h, markers1, markers2, merge1=NULL, merge2=NULL,
     model='additive', verbose=FALSE, family='gaussian' )
     epistasispair( h, marker1, marker2, merge1=NULL, merge2=NULL,
     model='additive', verbose=FALSE, d1=NULL, d2=NULL, main1=0, main2=0, family='gaussian' )

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

       h: an object returned by a previous call to happy()

markers1: an array of marker names or indices

markers2: an array of marker names or indices

 marker1: a single marker name or index

 marker2: a single marker name or index

  merge1: an optional merge object (returned by mergematrices())
          determining how the strains should be merged together for the
          markers listed in marker1

  merge2: an optional merge object (returned by mergematrices())
          determining how the strains should be merged together for the
          markers listed in marker2

   model: the type of model fitted at each locus. Either 'additive' or
          'full'

 verbose: switch controlling output to screen

      d1: optional design matrix for the main effect of the first
          marker (saves computation time)

   main1: optional log-P-value for the main effect of the first marker.
          NOTE: If d1 is not NULL then main1 _must_ be set 

      d2: optional design matrix for the main effect of the second 
          marker (saves computation time). 

   main2: optional log-P-value for the main effect of the second
          marker. NOTE: If d2 is not NULL then main2 _must_ be set 

  family: The distribution of errors in the data. The default is
          'gaussian'. This variable controls the type of model fitting.
          In the gauusian case a standard linear model is fitted using
          lm(). Otherwise the data are fitted as a generalised linear
          model using glm(), when the value of family must be one of
          the distributions hangled by glm(), such as 'binomial',
          'gamma'. See family() for the full range of models.

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

     epistasis() returns a matrix with columns named 'marker1',
     'marker2', 'main1', 'main2', 'main1+main2', 'main1*main2',
     'main1.main2'. marker1 and marker2 are the names of the markers
     being compared in a given row, the remaining values are the  ANOVA
     log-P-values of the main effects (main1 and main2), the combined
     additive effect (main1+main2), the addtive plus interaction
     (main1*main2) and the partial F of the interaction (main1.main2)
     after allowing for main1+main2. epistasispair() returns a list
     with the same fields.

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

     Richard Mott

