Analysing rearrangement breakpoint distributions by means of binomial confidence intervals

Non-randomness in the distribution of rearrangement breakpoints can be assessed by means of a binomial probability model. Using the binomial model, we compute confidence intervals for the observed number of breakpoints in chromosome bands and use these confidence intervals to test for non-randomness in a set of 452 constitutional rearrangement breakpoints. This method is useful because it describes the observed data and is not dependent on comparison with a pre-existing hypothesis regarding the distribution.