Individual haplotyping prediction agreements

Haplotype problems find important applications in various fields of human genetics. Modern sequencing technologies have significantly improved accuracy of reconstruction using individual's DNA short fragments. The objective of individual haplotype reconstruction algorithms is to design algorithms that rebuild the most likely copies of each chromosome from a set of given short fragments. The performance of the haplotype algorithms are often measured by comparing the reconstructed haplotypes with the original short fragments, using Minimum Error Correction measurement. Yet, there is no evidence in the literature about the level of agreement in all the solutions to the individual haplotyping problem. In other words, it is unknown if different haplotype algorithms coincide on the reconstructed/predicted haplotypes. In this paper, we investigate the problem of haplotyping prediction agreement by comparing the haplotypes produced by a number of well-known haplotype reconstruction algorithms.