GroupSim_pairwise_BMA

GroupSim_pairwise_BMA(
  dag,
  group1,
  group2,
  term_sim_method = NULL,
  IC_method = NULL,
  verbose = simona_opt$verbose,
  ...
)

Methods

BMA stands for "best-match average". First define similarity of a term to a group of terms as

S(x, group) = max_{y in group}(x, y)

which is the most similar terms in group to x.

Then the BMA similarity is calculated as:

group_sim = 0.5*(mean_{a in group1}(S(a, group2)) + mean_{b in group2}(S(b, group1)))

So it is the average of the similarity of every term in group1 to the whole group2 and every term in group2 to the whole group1.

The term semantic similarity method and the IC method can be set via control argument:

group_sim(dag, group1, group2, method = "GroupSim_pairwise_BMA"
    control = list(term_sim_method = "Sim_Lin_1998", IC_method = "IC_annotation")`.

Other parameters for the term_sim_method can also be set in the control list.