Improving estimates of the breakpoints in genome copy number alteration profiles with confidence masks

Chromosomal structural changes in human body known as copy number alterations (CNAs) are often associated with disease such as cancer. Therefore, accurate estimation of the CNAs using high resolution technologies is on a front line of bioinformatics and engineering. Since the Laplace distribution recently justified to represent jitter in the CNA breakpoints is not sufficiently accurate to estimate small changes, we propose a more accurate approximation based on the modified Bessel function of the second kind and zeroth order. We develop the relevant confidence masks to bound the CNA estimates for the given probability. The masks are applied to test the estimates obtained using the profile copy number of single nucleotide polymorphism (SNP) array data. We also show how to improve the estimates for the required confidence probability by removing some unlikely existing breakpoints.