
OFC = observed fold-change
Event A = mean of N fold-changes >= OFC
Event B = null hypothesis is true
Null hypothesis (Wilcoxon): P(count2 > count1) = P(count2 < count1)

P-value = P(A|B) = P(A ^ B) / P(B)

Solution:

S1 = { All FCs count m / count n and count n / count m : m != n }
   = { FCs of #samples choose 2 and their inverses }
   Need all fold-changes and their inverses to construct set of all
   fold-change means of N samples?

S2 = { Means of all subsets of N fold-changes from S1 }
   = { Means of { S1 choose N } }

For S2:

    MFC = mean of 1 subset of N fold-changes

	   |{ MFC >= OFC }|
    P(A) = ----------------
		 |S2|
    P(B) = 1
    P(A ^ B) = P(A)
    P(A|B) = P(A ^ B) / P(B)
	   = P(A) / 1
	   = P(A)

