An Algorithm for Bounding Error in Estimates of Genome Copy Number Variations Using SNP Array Technology

Abstract

Noise strongly affects measurements of chromosomal changes provided using the modern single nucleotide polymorphism (SNP) array technology. This makes difficulties in the estimation of genome copy number variations (CNVs) essential for human life. We propose an efficient algorithm for computing the confidence upper and lower boundary limits in order to guarantee an existence of genomic changes with a required probability. The algorithm is designed to approximate the breakpoint jitter probability with the discrete skew Laplace distribution. We test some SNP-based measurements by the upper and lower confidence bound masks and show special cases when the estimated chromosomal change may not exist and the breakpoint locations cannot be estimated with sufficient accuracy.