Number of significant positive selection aggregation tests with different sliding window size, with and without FDR correction, and using different BEB posterior probability cutoffs.