Patterns of homozygosity in patients with uniparental disomy: detection rate and suggested reporting thresholds for SNP microarrays

Abstract

Purpose: Single-nucleotide polymorphism (SNP) microarrays can easily identify whole-chromosome isodisomy but are unable to detect whole-chromosome heterodisomy. However, most cases of uniparental disomy (UPD) involve combinations of heterodisomy and isodisomy, visualized on SNP microarrays as long continuous stretches of homozygosity (LCSH). LCSH raise suspicion for, but are not diagnostic of, UPD, and reporting necessitates confirmatory testing. The goal of this study was to define optimal LCSH reporting standards. Methods: Eighty-nine individuals with known UPD were analyzed using chromosomal microarray. The LCSH patterns were compared with those in a phenotypically normal population to predict the clinical impact of various reporting thresholds. False-positive and -negative rates were calculated at various LCSH thresholds. Results: Twenty-seven of 84 cases with UPD had no significant LCSH on the involved chromosome. Fifty UPD-positive samples had LCSH of varying sizes: the average size of terminal LCSH was 11.0 megabases while the average size of interstitial LCSH was 24.1 megabases. LCSH in the normal population tended to be much smaller (average 4.3 megabases) and almost exclusively interstitial; however, overlap between the populations was noted. Conclusion: We hope that this work will aid clinical laboratories in the recognition and reporting of LCSH.