Genetic variation detection using maximum likelihood estimator

In recent years it has come to be appreciated that submicroscopic DNA copy number differences represent an important source of human genetic variation and contribute significantly to disease susceptibility. Array comparative genomic hybridization has emerged as a powerful tool for assessing copy number change and a number of algorithms have been developed to accurately assign copy number segments while minimizing errors from this inherently variable methodology. In this paper, we present an extended version of our previously proposed algorithm, maximum likelihood estimator, to clearly map and detect copy number variations. The extension accounts for both the unequal spacing distance between the contiguous probes and the regional evaluation of the detected segments based on biological information of the genomic positions. Using genomic DNA from well-characterized cell lines, we compare the performance of the proposed methods. Finally, the experimental results show that our proposed method outperforms other popular commercial programs and published algorithms.