A method for determining significant interactions. (**A**) The sequence of captured fragments is mapped to the genome, and the number of reads mapping to each 3C fragment is used as the observed distribution. (**B**) (Step 1) Sliding windows of a desired number of 3C fragments, *W*, are demarcated, and the total reads in each window are determined for the observed data. (Step 2) The reads on the chromosome are distributed randomly among 3C fragments and a new reads per window is calculated. (Step 3) A cutoff, *X*, is calculated for a desired FDR using the shuffled reads per window data. (Step 4) *X* is calculated from at least 1000 random shuffles to generate a histogram of possible cutoffs for the desired FDR. (Step 5) The final threshold for calling a significantly enriched window in the observed data set is set at the 95th percentile for calculated *X*s. In the depicted example, a desired FDR of < 0.01 leads to a final threshold of 40 reads per window for calling significant interactions in the 4C data. (**C**) Ten significance thresholds were calculated independently using an FDR cutoff based on average randomly permutated distributions (left) and a cutoff determined from confidence bounds applied to a distribution of FDR calculations (right). The y-axis represents significance thresholds in reads per window. Both calculations represent an FDR of <0.01.

## PubMed Commons