We can calculate the probability that a given random pairwise similarity score X is bigger than a value x as p(X > x) using the fitted Gaussian function, we can transform a Tanimoto similarity matrix into a p-value p(X > x) as follows:
were t(xi,xj) is the tanimoto similarity matrix and h is the smoothing factor which you need to estimate.
Hope now you all can very easily understand how you can calculate your pvalue from a large distribution.