Representational oligonucleotide microarray analysis (ROMA) and comparison of binning and change-point methods of analysis: application to detection of del22q11.2 (DiGeorge) Syndrome

Abstract

DiGeorge (del22q11.2) syndrome is estimated to occur in 1:4,000 births, is the most common contiguous-gene deletion syndrome in humans, and is caused by autosomal dominant deletions in the 22q11.2 DiGeorge syndrome critical region (DGCR). Multiple microarray methods have been developed recently for analyzing such copy number changes, but data analysis and accurate deletion detection remains challenging. Clinical use of these microarray methods would have many advantages, particularly when the possibility of a chromosomal disorder cannot be determined simply on the basis of history and physical examination data alone. We investigated the use of the microarray technique, representational oligonucleotide microarray analysis (ROMA), in the detection of del22q11.2 syndrome. Genomic DNA was isolated from three well-characterized cell lines with 22q11.2 DGCR deletions and from the blood of a patient suspected of having del22q11.2 syndrome, and analyzed using both the binning and change-point model algorithms. Though the 22q11.2 deletion was easily identified with either method, change-point models provide clearer identification of deleted regions, with the potential for fewer false-positive results. For circumstances in which a clear, a priori, copy-number change hypothesis is not present, such as in many clinical samples, change-point methods of analysis may be easier to interpret.