We call a “configuration” one possible combination of pairs of binary vectors indicating whether the variant is associated with the selected trait. We can group the configurations into five sets, , , , , , containing assignments of all SNPs Q to the functional role corresponding to the five hypothesis , , , , . We can compute the posterior probabilities given the data for each of these 5 hypothesis by summing over the relevant configurations:(1)where P(S) is the prior probability of a configuration, is the probability of the observed data D given a configuration S, and the sum is over all configurations S which are consistent with a given hypothesis , where h = (1,2,3,4). Thus, the probability of the data given a configuration is weighted by the prior probability of that configuration.
